Skip to content

[WIP] Simplifying refactor #74

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom
Draft
hunhoffe wants to merge 100 commits into
base: devel
Choose a base branch
from
Draft

[WIP] Simplifying refactor #74

hunhoffe wants to merge 100 commits into from

Conversation

@hunhoffe
Copy link
Collaborator

@hunhoffe hunhoffe commented Feb 5, 2026

The draftiest of draft PRs.

Added

Changed

Removed

PR Merge Checklist

  1. The PR is rebased on the latest devel commit and pointing to devel.
  2. Your PR has been reviewed and approved.
  3. All checks are passing.

andrej added 30 commits January 14, 2026 14:46
@andrej
rework profiling
cf6485d
@andrej
vibe-coded flame graph visualization
af46d81
@andrej
plot updates
1f184eb
@andrej
simplified implementation (no KV cache yet)
eedc527
@andrej
add KV cache
5372fca
@andrej
start with simplified llama for NPU
f0d3289
@andrej
offload last layer GEMM/GEMV
fd6f759
@andrej
add profile path analyzer
ec0d372
@andrej
change profiling to allow annotation using contexts; remove nn.Module
280917a
@andrej
refactoring started; RMSNorm offloaded, last layer GEMM started
7b56ef8
@andrej
last layer GEMM offloaded
1daaa39
@andrej
fixes:
83380c2
@andrej
cleanup
3aa1e02
@andrej
RMSNorm offloaded verywhere, cleanup
002d487
@andrej
less buffer copying
db9c4b8
@andrej
offload first residual
b0942e8
@andrej
simplify
02e5307
@andrej
offload second residual
e815c50
@andrej
SwiGLU offloaded
ebe8f32
@andrej
offload RoPE
7515199
@andrej
offload attention query projection linear layer
7ef6b03
@andrej
offload attention key projection linear layer -- slight decrease in o…
d43eeea 
…utput quality
@andrej
add batching to GEMV, fix issue when K<vector_size, offload attention…
de2995c 
… score calculation in llama
@andrej
add strided_copy operator
74300bd
@andrej
add patchable callable
b1eab7c
@andrej
simplify llama_npu.py, make GEMV operator input shapes simpler
119bb7e
@andrej
fix strided copy; offload KV cache concat + transpose to NPU
7fee60d
@andrej
offload repeat_interleave
b72432b
@andrej
offload normalization/scaling + softmax (with -inf masking on CPU for…
8b4b2b3 
… now)
@andrej
make softmax run-time parametrizable
5575d4f
@github-actions
Copy link
Contributor

github-actions bot commented Feb 6, 2026

📊 Test Results for Small Benchmark/Test Suite

e40f553 (2026_02_06_23_11_00)

IRONCLAD

Tested on 2026_02_06_23_11_00 at commit e40f553.

Test Checks Latency (mean)Bandwidth (mean)GFLOP/s (mean)
axpy_1_cols_2_channels_2048_tile_2048_3.0_0 ✅ 5/5 157.54 0.08 n/a
axpy_2_cols_2_channels_2048_tile_1024_3.0_0 ✅ 5/5 163.00 0.08 n/a
axpy_4_cols_2_channels_2048_tile_512_3.0_0 ✅ 5/5 176.32 0.07 n/a
axpy_8_cols_2_channels_2048_tile_256_3.0_0 ✅ 5/5 221.30 0.06 n/a
dequant_1_cols_1_channels_2048_tile_2048_0 ✅ 5/5 154.02 0.04 n/a
dequant_1_cols_2_channels_2048_tile_1024_0 ✅ 5/5 120.86 0.04 n/a
dequant_2_cols_1_channels_2048_tile_1024_0 ✅ 5/5 170.82 0.03 n/a
dequant_2_cols_2_channels_2048_tile_512_0 ✅ 5/5 151.72 0.04 n/a
dequant_4_cols_1_channels_2048_tile_512_0 ✅ 5/5 163.36 0.03 n/a
dequant_4_cols_2_channels_2048_tile_256_0 ✅ 5/5 157.86 0.03 n/a
dequant_8_cols_1_channels_2048_tile_256_0 ✅ 5/5 160.08 0.03 n/a
dequant_8_cols_2_channels_2048_tile_128_0 ✅ 5/5 196.14 0.03 n/a
eltwise_add_1_cols_2_channels_2048_tile_2048 ✅ 5/5 161.64 0.08 n/a
eltwise_add_2_cols_2_channels_2048_tile_1024 ✅ 5/5 175.56 0.07 n/a
eltwise_add_4_cols_2_channels_2048_tile_512 ✅ 5/5 144.40 0.09 n/a
eltwise_add_8_cols_2_channels_2048_tile_256 ✅ 5/5 189.06 0.07 n/a
eltwise_mul_1_cols_2_channels_2048_tile_2048 ✅ 5/5 129.70 0.10 n/a
eltwise_mul_2_cols_2_channels_2048_tile_1024 ✅ 5/5 131.94 0.09 n/a
eltwise_mul_4_cols_2_channels_2048_tile_512 ✅ 5/5 161.74 0.08 n/a
eltwise_mul_8_cols_2_channels_2048_tile_256 ✅ 5/5 214.32 0.07 n/a
gelu_1_cols_1_channels_2048_tile_2048 ✅ 5/5 150.58 0.06 n/a
gelu_1_cols_2_channels_2048_tile_1024 ✅ 5/5 154.06 0.06 n/a
gelu_2_cols_1_channels_2048_tile_1024 ✅ 5/5 160.78 0.05 n/a
gelu_2_cols_2_channels_2048_tile_512 ✅ 5/5 166.78 0.05 n/a
gelu_4_cols_1_channels_2048_tile_512 ✅ 5/5 184.12 0.05 n/a
gelu_4_cols_2_channels_2048_tile_256 ✅ 5/5 149.50 0.06 n/a
gelu_8_cols_1_channels_2048_tile_256 ✅ 5/5 178.92 0.05 n/a
gelu_8_cols_2_channels_2048_tile_128 ✅ 5/5 211.24 0.04 n/a
gemm_1792x896x1152_64x32x48_8cols_ccolmaj ✅ 5/5 2175.70 4.35 n/a
gemm_192x384x64_48x96x16_4cols ✅ 5/5 202.20 1.11 n/a
gemm_192x384x64_48x96x16_4cols_bcolmaj_ccolmaj ✅ 5/5 236.88 1.05 n/a
gemm_2048x2048x2048_64x64x64_1cols ✅ 5/5 48472.28 0.52 n/a
gemm_2048x2048x2048_64x64x64_2cols_bcolmaj ✅ 5/5 27817.50 0.90 n/a
gemm_2048x2048x2048_64x64x64_8cols_bcolmaj_ccolmaj ✅ 5/5 7846.12 3.21 n/a
gemm_384x1536x1792_32x48x64_4cols_bcolmaj ✅ 5/5 2190.02 3.69 n/a
gemm_896x1792x640_32x64x80_8cols_ccolmaj ✅ 5/5 1285.16 5.31 n/a
layer_norm_1_cols_1_channels_2048_tile_2048 ✅ 5/5 189.96 0.04 n/a
layer_norm_1_cols_2_channels_2048_tile_1024 ✅ 5/5 181.42 0.05 n/a
layer_norm_2_cols_1_channels_2048_tile_1024 ✅ 5/5 226.30 0.04 n/a
layer_norm_2_cols_2_channels_2048_tile_512 ✅ 5/5 179.32 0.05 n/a
layer_norm_4_cols_1_channels_2048_tile_512 ✅ 5/5 195.22 0.04 n/a
layer_norm_4_cols_2_channels_2048_tile_256 ✅ 5/5 222.74 0.04 n/a
layer_norm_8_cols_1_channels_2048_tile_256 ✅ 5/5 237.82 0.04 n/a
layer_norm_8_cols_2_channels_2048_tile_128 ✅ 5/5 242.14 0.03 n/a
matrix_vector_mul_128x128_32tsi_128tso_1col0 ✅ 5/5 n/a 0.21 n/a
matrix_vector_mul_2048x8192_1tsi_1024tso_2col0 ✅ 5/5 n/a 24.50 n/a
matrix_vector_mul_2048x8192_1tsi_2048tso_1col0 ✅ 5/5 n/a 12.52 n/a
matrix_vector_mul_2048x8192_1tsi_256tso_8col0 ✅ 5/5 n/a 41.74 n/a
matrix_vector_mul_2048x8192_1tsi_512tso_4col0 ✅ 5/5 n/a 41.17 n/a
matrix_vector_mul_8192x2048_4tsi_1024tso_1col0 ✅ 5/5 n/a 12.96 n/a
matrix_vector_mul_8192x2048_4tsi_1024tso_2col0 ✅ 5/5 n/a 24.42 n/a
matrix_vector_mul_8192x2048_4tsi_1024tso_4col0 ✅ 5/5 n/a 41.32 n/a
matrix_vector_mul_8192x2048_4tsi_1024tso_8col0 ✅ 5/5 n/a 43.47 n/a
mem_copy_16_cores_2_chans_2048_tile_128_False0 ✅ 5/5 212.10 0.04 n/a
mem_copy_1_cores_1_chans_2048_tile_2048_False0 ✅ 5/5 178.74 0.05 n/a
mem_copy_2_cores_1_chans_2048_tile_1024_False0 ✅ 5/5 190.50 0.04 n/a
mem_copy_2_cores_2_chans_2048_tile_1024_False0 ✅ 5/5 179.38 0.05 n/a
mem_copy_4_cores_1_chans_2048_tile_512_False0 ✅ 5/5 166.66 0.05 n/a
mem_copy_4_cores_2_chans_2048_tile_512_False0 ✅ 5/5 199.98 0.04 n/a
mem_copy_8_cores_1_chans_2048_tile_256_False0 ✅ 5/5 186.66 0.05 n/a
mem_copy_8_cores_2_chans_2048_tile_256_False0 ✅ 5/5 182.84 0.05 n/a
mha0 ✅ 5/5 47526.30 0.18 n/a
relu_1_cols_1_channels_2048_tile_2048 ✅ 5/5 157.60 0.05 n/a
relu_2_cols_1_channels_2048_tile_1024 ✅ 5/5 173.92 0.05 n/a
relu_4_cols_1_channels_2048_tile_512 ✅ 5/5 173.02 0.05 n/a
relu_8_cols_1_channels_2048_tile_256 ✅ 5/5 168.88 0.05 n/a
rms_norm_1_cols_1_channels_2048_tile_2048 ✅ 5/5 165.92 0.05 n/a
rms_norm_1_cols_2_channels_2048_tile_1024 ✅ 5/5 167.80 0.05 n/a
rms_norm_2_cols_1_channels_2048_tile_1024 ✅ 5/5 156.94 0.05 n/a
rms_norm_2_cols_2_channels_2048_tile_512 ✅ 5/5 168.50 0.05 n/a
rms_norm_4_cols_1_channels_2048_tile_512 ✅ 5/5 188.74 0.04 n/a
rms_norm_4_cols_2_channels_2048_tile_256 ✅ 5/5 173.28 0.05 n/a
rms_norm_8_cols_1_channels_2048_tile_256 ✅ 5/5 190.52 0.05 n/a
rms_norm_8_cols_2_channels_2048_tile_128 ✅ 5/5 212.72 0.04 n/a
rope_1c_32rows_512cols_32arows_0m ✅ 5/5 157.08 0.63 n/a
rope_1c_32rows_512cols_8arows_0m ✅ 5/5 150.52 0.50 n/a
rope_2c_32rows_512cols_32arows_0m ✅ 5/5 158.82 0.64 n/a
rope_2c_32rows_512cols_8arows_0m ✅ 5/5 140.98 0.53 n/a
rope_8c_32rows_512cols_32arows_0m ✅ 5/5 172.44 0.58 n/a
rope_8c_32rows_512cols_8arows_0m ✅ 5/5 190.32 0.41 n/a
sigmoid_1_cols_1_channels_2048_tile_2048 ✅ 5/5 142.34 0.06 n/a
sigmoid_2_cols_1_channels_2048_tile_1024 ✅ 5/5 148.20 0.06 n/a
sigmoid_4_cols_1_channels_2048_tile_512 ✅ 5/5 145.34 0.06 n/a
sigmoid_8_cols_1_channels_2048_tile_256 ✅ 5/5 161.46 0.05 n/a
silu_1_cols_1_channels_2048_tile_2048 ✅ 5/5 164.72 0.05 n/a
silu_2_cols_1_channels_2048_tile_1024 ✅ 5/5 174.14 0.05 n/a
silu_4_cols_1_channels_2048_tile_512 ✅ 5/5 153.50 0.06 n/a
silu_8_cols_1_channels_2048_tile_256 ✅ 5/5 199.18 0.04 n/a
softmax_2_cols_2_channels_32768_tile_1024 ✅ 5/5 180.34 0.76 n/a
softmax_2_cols_2_channels_32768_tile_512 ✅ 5/5 158.50 0.85 n/a
softmax_4_cols_4_channels_32768_tile_2048 ✅ 5/5 257.22 0.57 n/a
swiglu_decode_1x2048x2048_0 ✅ 5/5 n/a n/a n/a
swiglu_prefill_256x2048x2048_0 ❌ 0/5 n/a n/a n/a
tanh_1_cols_1_channels_2048_tile_2048 ✅ 5/5 169.04 0.05 n/a
tanh_2_cols_1_channels_2048_tile_1024 ✅ 5/5 154.10 0.06 n/a
tanh_4_cols_1_channels_2048_tile_512 ✅ 5/5 173.00 0.05 n/a
tanh_8_cols_1_channels_2048_tile_256 ✅ 5/5 172.88 0.05 n/a
transpose_2048_M_64_N_1_cols_1_channels_64_m_64_n_8_s0 ✅ 5/5 193.46 2.85 n/a
transpose_2048_M_64_N_1_cols_2_channels_64_m_64_n_8_s0 ✅ 5/5 177.18 3.13 n/a
weighted_rms_norm_1_cols_2_channels_2048_weights_2048 ✅ 5/5 165.86 0.08 n/a
weighted_rms_norm_2_cols_2_channels_2048_weights_1024 ✅ 5/5 160.38 0.07 n/a
weighted_rms_norm_4_cols_2_channels_2048_weights_512 ✅ 5/5 181.10 0.05 n/a
weighted_rms_norm_8_cols_2_channels_2048_weights_256 ✅ 5/5 247.62 0.04 n/a
📈 Trends (vs main branch) for Small Benchmark/Test Suite

e40f553 (2026_02_06_23_11_00)

IRONCLAD Trends

axpy_1_cols_2_channels_2048_tile_2048_3.0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.13 (-1.96%)0.11 (+2.52%)0.10 (+1.86%)0.08 (-1.91%)0.02 (-1.72%)162.40 (+1.95%)120.96 (-2.36%)117.00 (-1.85%)94.40 (+2.05%)23.11 (+4.09%)
0a6c11c — 2025-12-04 00:39:100.13 (n/a)0.10 (n/a)0.10 (n/a)0.08 (n/a)0.02 (n/a)159.30 (n/a)123.88 (n/a)119.20 (n/a)92.50 (n/a)22.21 (n/a)

axpy_1_cols_2_channels_2048_tile_2048_3.0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.10 (-38.39%)0.08 (-36.55%)0.08 (-30.24%)0.06 (-43.45%)0.02 (-44.03%)210.00 (+76.77%)157.54 (+57.07%)158.30 (+43.39%)126.60 (+62.31%)33.31 (+62.56%)
1531781 — 2026-01-23 18:49:240.16 (n/a)0.13 (n/a)0.11 (n/a)0.10 (n/a)0.03 (n/a)118.80 (n/a)100.30 (n/a)110.40 (n/a)78.00 (n/a)20.49 (n/a)

axpy_2_cols_2_channels_2048_tile_1024_3.0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.11 (-26.77%)0.09 (-9.09%)0.09 (-10.81%)0.08 (+8.25%)0.01 (-55.42%)152.00 (-7.60%)135.46 (+6.28%)142.80 (+12.18%)115.00 (+36.58%)15.03 (-41.36%)
0a6c11c — 2025-12-04 00:39:100.15 (n/a)0.10 (n/a)0.10 (n/a)0.07 (n/a)0.02 (n/a)164.50 (n/a)127.46 (n/a)127.30 (n/a)84.20 (n/a)25.64 (n/a)

axpy_2_cols_2_channels_2048_tile_1024_3.0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.10 (-30.92%)0.08 (-35.89%)0.09 (-37.01%)0.05 (-32.16%)0.02 (-30.38%)231.30 (+47.42%)163.00 (+56.04%)143.30 (+58.69%)127.30 (+44.82%)42.96 (+45.55%)
1531781 — 2026-01-23 18:49:240.14 (n/a)0.12 (n/a)0.14 (n/a)0.08 (n/a)0.03 (n/a)156.90 (n/a)104.46 (n/a)90.30 (n/a)87.90 (n/a)29.52 (n/a)

axpy_4_cols_2_channels_2048_tile_512_3.0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.12 (-10.21%)0.10 (-8.41%)0.10 (-6.20%)0.07 (+3.52%)0.02 (-23.48%)166.70 (-3.36%)128.78 (+7.33%)121.80 (+6.65%)103.90 (+11.36%)22.95 (-18.82%)
0a6c11c — 2025-12-04 00:39:100.13 (n/a)0.11 (n/a)0.11 (n/a)0.07 (n/a)0.02 (n/a)172.50 (n/a)119.98 (n/a)114.20 (n/a)93.30 (n/a)28.27 (n/a)

axpy_4_cols_2_channels_2048_tile_512_3.0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.08 (-45.94%)0.07 (-48.89%)0.07 (-53.97%)0.06 (-34.38%)0.01 (-64.13%)204.30 (+52.35%)176.32 (+90.74%)181.20 (+117.27%)145.40 (+84.99%)22.67 (-3.06%)
1531781 — 2026-01-23 18:49:240.16 (n/a)0.14 (n/a)0.15 (n/a)0.09 (n/a)0.03 (n/a)134.10 (n/a)92.44 (n/a)83.40 (n/a)78.60 (n/a)23.38 (n/a)

axpy_8_cols_2_channels_2048_tile_256_3.0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.10 (-16.19%)0.08 (-10.40%)0.09 (-2.09%)0.07 (-17.38%)0.01 (-10.61%)184.40 (+21.00%)153.18 (+11.99%)144.10 (+2.20%)127.10 (+19.34%)21.26 (+34.76%)
0a6c11c — 2025-12-04 00:39:100.12 (n/a)0.09 (n/a)0.09 (n/a)0.08 (n/a)0.01 (n/a)152.40 (n/a)136.78 (n/a)141.00 (n/a)106.50 (n/a)15.78 (n/a)

axpy_8_cols_2_channels_2048_tile_256_3.0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.07 (-57.08%)0.06 (-58.57%)0.06 (-60.96%)0.04 (-55.72%)0.01 (-57.51%)331.60 (+125.89%)221.30 (+139.92%)204.80 (+156.00%)168.60 (+133.20%)66.79 (+114.98%)
1531781 — 2026-01-23 18:49:240.17 (n/a)0.14 (n/a)0.15 (n/a)0.08 (n/a)0.03 (n/a)146.80 (n/a)92.24 (n/a)80.00 (n/a)72.30 (n/a)31.07 (n/a)

dequant_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.05 (-18.83%)0.04 (-5.21%)0.04 (-2.69%)0.03 (+37.10%)0.00 (-53.98%)160.40 (-27.06%)141.12 (-0.91%)136.40 (+2.79%)114.30 (+23.17%)17.37 (-59.06%)
0a6c11c — 2025-12-04 00:39:100.06 (n/a)0.04 (n/a)0.04 (n/a)0.02 (n/a)0.01 (n/a)219.90 (n/a)142.42 (n/a)132.70 (n/a)92.80 (n/a)42.43 (n/a)

dequant_1_cols_1_channels_2048_tile_2048_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.05 (-21.35%)0.04 (-24.06%)0.03 (-26.51%)0.03 (-31.92%)0.01 (+2.11%)184.30 (+46.85%)154.02 (+33.84%)165.50 (+36.10%)112.10 (+27.24%)29.76 (+93.26%)
1531781 — 2026-01-23 18:49:240.06 (n/a)0.05 (n/a)0.04 (n/a)0.04 (n/a)0.01 (n/a)125.50 (n/a)115.08 (n/a)121.60 (n/a)88.10 (n/a)15.40 (n/a)

dequant_1_cols_2_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.06 (-13.28%)0.05 (-8.93%)0.06 (-13.59%)0.04 (+4.63%)0.01 (-39.10%)133.60 (-4.43%)106.08 (+5.85%)95.10 (+15.69%)87.90 (+15.35%)17.90 (-31.97%)
0a6c11c — 2025-12-04 00:39:100.07 (n/a)0.06 (n/a)0.06 (n/a)0.04 (n/a)0.01 (n/a)139.80 (n/a)100.22 (n/a)82.20 (n/a)76.20 (n/a)26.31 (n/a)

dequant_1_cols_2_channels_2048_tile_1024_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.05 (-33.56%)0.04 (-10.37%)0.04 (-7.36%)0.04 (+24.64%)0.00 (-82.95%)131.40 (-19.73%)120.86 (+5.41%)120.20 (+8.00%)115.20 (+50.59%)6.49 (-79.47%)
1531781 — 2026-01-23 18:49:240.07 (n/a)0.05 (n/a)0.05 (n/a)0.03 (n/a)0.01 (n/a)163.70 (n/a)114.66 (n/a)111.30 (n/a)76.50 (n/a)31.63 (n/a)

dequant_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.07 (+5.66%)0.05 (+1.21%)0.05 (+0.61%)0.04 (-15.02%)0.01 (+58.42%)143.40 (+17.64%)101.02 (+1.49%)97.60 (-0.61%)78.90 (-5.40%)22.56 (+78.52%)
0a6c11c — 2025-12-04 00:39:100.06 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)121.90 (n/a)99.54 (n/a)98.20 (n/a)83.40 (n/a)12.64 (n/a)

dequant_2_cols_1_channels_2048_tile_1024_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.04 (-23.20%)0.03 (-33.93%)0.03 (-40.14%)0.03 (-22.82%)0.00 (-26.72%)191.50 (+29.57%)170.82 (+50.98%)177.90 (+67.04%)132.10 (+30.28%)23.39 (+19.25%)
1531781 — 2026-01-23 18:49:240.05 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)147.80 (n/a)113.14 (n/a)106.50 (n/a)101.40 (n/a)19.61 (n/a)

dequant_2_cols_2_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.05 (+17.06%)0.04 (+8.98%)0.05 (+21.37%)0.03 (-11.65%)0.01 (+301.89%)159.70 (+13.18%)126.32 (-5.68%)109.00 (-17.67%)106.80 (-14.56%)22.48 (+280.15%)
0a6c11c — 2025-12-04 00:39:100.04 (n/a)0.04 (n/a)0.04 (n/a)0.04 (n/a)0.00 (n/a)141.10 (n/a)133.92 (n/a)132.40 (n/a)125.00 (n/a)5.91 (n/a)

dequant_2_cols_2_channels_2048_tile_512_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.05 (-14.54%)0.04 (-19.43%)0.04 (-27.58%)0.02 (-25.87%)0.01 (-13.24%)213.60 (+34.93%)151.72 (+25.00%)146.60 (+38.04%)114.90 (+17.01%)37.13 (+42.12%)
1531781 — 2026-01-23 18:49:240.05 (n/a)0.04 (n/a)0.05 (n/a)0.03 (n/a)0.01 (n/a)158.30 (n/a)121.38 (n/a)106.20 (n/a)98.20 (n/a)26.13 (n/a)

dequant_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.06 (+20.24%)0.05 (-3.19%)0.05 (-6.73%)0.03 (-20.69%)0.01 (+194.26%)156.10 (+26.09%)120.86 (+7.15%)114.90 (+7.18%)88.30 (-16.85%)24.40 (+214.82%)
0a6c11c — 2025-12-04 00:39:100.05 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.00 (n/a)123.80 (n/a)112.80 (n/a)107.20 (n/a)106.20 (n/a)7.75 (n/a)

dequant_4_cols_1_channels_2048_tile_512_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.05 (-24.63%)0.03 (-30.89%)0.03 (-34.39%)0.02 (-23.47%)0.01 (-28.59%)215.30 (+30.64%)163.36 (+43.78%)157.40 (+52.52%)112.50 (+32.67%)40.47 (+24.33%)
1531781 — 2026-01-23 18:49:240.06 (n/a)0.05 (n/a)0.05 (n/a)0.03 (n/a)0.01 (n/a)164.80 (n/a)113.62 (n/a)103.20 (n/a)84.80 (n/a)32.55 (n/a)

dequant_4_cols_2_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.05 (-6.09%)0.04 (-8.04%)0.04 (-12.46%)0.03 (-0.41%)0.01 (-20.19%)168.10 (+0.42%)135.74 (+7.44%)132.20 (+14.26%)107.40 (+6.44%)22.91 (-12.67%)
0a6c11c — 2025-12-04 00:39:100.05 (n/a)0.04 (n/a)0.05 (n/a)0.03 (n/a)0.01 (n/a)167.40 (n/a)126.34 (n/a)115.70 (n/a)100.90 (n/a)26.23 (n/a)

dequant_4_cols_2_channels_2048_tile_256_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.04 (-42.60%)0.03 (-34.00%)0.03 (-40.45%)0.03 (+1.69%)0.00 (-74.43%)185.30 (-1.65%)157.86 (+40.05%)157.60 (+68.02%)138.80 (+74.15%)18.01 (-58.68%)
1531781 — 2026-01-23 18:49:240.07 (n/a)0.05 (n/a)0.06 (n/a)0.03 (n/a)0.01 (n/a)188.40 (n/a)112.72 (n/a)93.80 (n/a)79.70 (n/a)43.58 (n/a)

dequant_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.04 (-13.65%)0.04 (-13.67%)0.04 (-18.56%)0.03 (-13.52%)0.00 (-23.88%)170.40 (+15.68%)140.12 (+15.33%)135.80 (+22.78%)119.60 (+15.78%)17.94 (+1.87%)
0a6c11c — 2025-12-04 00:39:100.05 (n/a)0.04 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)147.30 (n/a)121.50 (n/a)110.60 (n/a)103.30 (n/a)17.61 (n/a)

dequant_8_cols_1_channels_2048_tile_256_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.04 (-10.49%)0.03 (-11.65%)0.03 (-11.22%)0.02 (-23.64%)0.01 (+32.43%)218.60 (+30.98%)160.08 (+15.45%)153.10 (+12.66%)131.40 (+11.73%)34.85 (+94.19%)
1531781 — 2026-01-23 18:49:240.04 (n/a)0.04 (n/a)0.04 (n/a)0.03 (n/a)0.00 (n/a)166.90 (n/a)138.66 (n/a)135.90 (n/a)117.60 (n/a)17.94 (n/a)

dequant_8_cols_2_channels_2048_tile_128

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.05 (+17.21%)0.03 (-1.70%)0.03 (-3.04%)0.03 (-16.20%)0.01 (+95.59%)198.20 (+19.33%)160.00 (+4.92%)166.20 (+3.17%)106.50 (-14.66%)29.86 (+87.19%)
0a6c11c — 2025-12-04 00:39:100.04 (n/a)0.03 (n/a)0.03 (n/a)0.03 (n/a)0.00 (n/a)166.10 (n/a)152.50 (n/a)161.10 (n/a)124.80 (n/a)15.95 (n/a)

dequant_8_cols_2_channels_2048_tile_128_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.03 (-35.84%)0.03 (-29.61%)0.03 (-31.92%)0.02 (-5.66%)0.00 (-61.32%)243.70 (+6.00%)196.14 (+34.71%)192.40 (+46.87%)165.50 (+55.84%)30.13 (-38.55%)
1531781 — 2026-01-23 18:49:240.05 (n/a)0.04 (n/a)0.04 (n/a)0.02 (n/a)0.01 (n/a)229.90 (n/a)145.60 (n/a)131.00 (n/a)106.20 (n/a)49.03 (n/a)

eltwise_add_1_cols_2_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.11 (-35.65%)0.08 (-34.96%)0.08 (-23.55%)0.06 (-25.98%)0.02 (-53.93%)210.50 (+35.11%)161.64 (+47.24%)157.60 (+30.79%)114.70 (+55.42%)34.76 (+1.11%)
1531781 — 2026-01-23 18:49:240.17 (n/a)0.12 (n/a)0.10 (n/a)0.08 (n/a)0.04 (n/a)155.80 (n/a)109.78 (n/a)120.50 (n/a)73.80 (n/a)34.38 (n/a)

eltwise_add_2_cols_2_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.11 (-26.92%)0.07 (-41.09%)0.07 (-48.60%)0.05 (-38.72%)0.02 (-12.51%)239.40 (+63.19%)175.56 (+73.48%)178.30 (+94.44%)111.30 (+36.73%)47.03 (+80.93%)
1531781 — 2026-01-23 18:49:240.15 (n/a)0.13 (n/a)0.13 (n/a)0.08 (n/a)0.03 (n/a)146.70 (n/a)101.20 (n/a)91.70 (n/a)81.40 (n/a)26.00 (n/a)

eltwise_add_4_cols_2_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.11 (-33.26%)0.09 (-39.31%)0.09 (-38.19%)0.06 (-39.01%)0.02 (-20.86%)190.30 (+64.05%)144.40 (+67.01%)130.90 (+61.80%)109.00 (+49.93%)33.06 (+91.93%)
1531781 — 2026-01-23 18:49:240.17 (n/a)0.15 (n/a)0.15 (n/a)0.11 (n/a)0.02 (n/a)116.00 (n/a)86.46 (n/a)80.90 (n/a)72.70 (n/a)17.22 (n/a)

eltwise_add_8_cols_2_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.08 (-51.63%)0.07 (-51.29%)0.07 (-48.57%)0.05 (-58.78%)0.01 (-30.68%)265.60 (+142.56%)189.06 (+110.07%)167.10 (+94.30%)155.70 (+106.77%)46.18 (+243.71%)
1531781 — 2026-01-23 18:49:240.16 (n/a)0.14 (n/a)0.14 (n/a)0.11 (n/a)0.02 (n/a)109.50 (n/a)90.00 (n/a)86.00 (n/a)75.30 (n/a)13.44 (n/a)

eltwise_mul_1_cols_2_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.14 (+0.07%)0.10 (-19.00%)0.10 (-27.68%)0.06 (-41.45%)0.03 (+128.86%)201.70 (+70.79%)129.70 (+32.51%)128.00 (+38.23%)90.40 (+0.00%)44.31 (+283.65%)
1531781 — 2026-01-23 18:49:240.14 (n/a)0.13 (n/a)0.13 (n/a)0.10 (n/a)0.01 (n/a)118.10 (n/a)97.88 (n/a)92.60 (n/a)90.40 (n/a)11.55 (n/a)

eltwise_mul_2_cols_2_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.10 (-22.81%)0.09 (-23.38%)0.10 (-25.85%)0.08 (-13.56%)0.01 (-46.06%)150.40 (+15.69%)131.94 (+29.23%)125.90 (+34.94%)118.00 (+29.53%)13.45 (-18.60%)
1531781 — 2026-01-23 18:49:240.13 (n/a)0.12 (n/a)0.13 (n/a)0.09 (n/a)0.02 (n/a)130.00 (n/a)102.10 (n/a)93.30 (n/a)91.10 (n/a)16.53 (n/a)

eltwise_mul_4_cols_2_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.12 (-27.63%)0.08 (-45.32%)0.07 (-54.08%)0.07 (-38.84%)0.02 (+2.03%)186.40 (+63.51%)161.74 (+87.68%)177.40 (+117.67%)102.50 (+38.14%)34.03 (+114.67%)
1531781 — 2026-01-23 18:49:240.17 (n/a)0.15 (n/a)0.15 (n/a)0.11 (n/a)0.02 (n/a)114.00 (n/a)86.18 (n/a)81.50 (n/a)74.20 (n/a)15.85 (n/a)

eltwise_mul_8_cols_2_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.10 (-19.56%)0.07 (-40.63%)0.06 (-45.65%)0.03 (-59.72%)0.03 (+39.00%)384.30 (+148.26%)214.32 (+90.88%)192.40 (+83.94%)118.70 (+24.29%)102.94 (+326.49%)
1531781 — 2026-01-23 18:49:240.13 (n/a)0.11 (n/a)0.12 (n/a)0.08 (n/a)0.02 (n/a)154.80 (n/a)112.28 (n/a)104.60 (n/a)95.50 (n/a)24.14 (n/a)

gelu_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.06 (-43.80%)0.06 (-44.90%)0.06 (-50.19%)0.04 (-45.35%)0.01 (-56.72%)186.00 (+83.07%)150.58 (+79.69%)145.40 (+100.83%)128.50 (+77.98%)22.51 (+42.91%)
1531781 — 2026-01-23 18:49:240.11 (n/a)0.10 (n/a)0.11 (n/a)0.08 (n/a)0.02 (n/a)101.60 (n/a)83.80 (n/a)72.40 (n/a)72.20 (n/a)15.75 (n/a)

gelu_1_cols_2_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.07 (-44.17%)0.06 (-47.96%)0.05 (-59.09%)0.04 (-42.42%)0.01 (-43.36%)191.00 (+73.64%)154.06 (+91.95%)168.80 (+144.28%)119.00 (+79.22%)31.74 (+71.25%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.11 (n/a)0.12 (n/a)0.07 (n/a)0.02 (n/a)110.00 (n/a)80.26 (n/a)69.10 (n/a)66.40 (n/a)18.54 (n/a)

gelu_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.05 (-54.67%)0.05 (-50.69%)0.05 (-55.57%)0.05 (-22.64%)0.00 (-84.66%)178.10 (+29.25%)160.78 (+89.11%)153.50 (+125.07%)149.50 (+120.50%)13.24 (-56.21%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.10 (n/a)0.12 (n/a)0.06 (n/a)0.03 (n/a)137.80 (n/a)85.02 (n/a)68.20 (n/a)67.80 (n/a)30.24 (n/a)

gelu_2_cols_2_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.06 (-48.76%)0.05 (-53.66%)0.05 (-56.60%)0.04 (-51.10%)0.01 (-43.92%)220.50 (+104.55%)166.78 (+117.16%)160.20 (+130.50%)130.90 (+95.08%)37.44 (+115.17%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.11 (n/a)0.12 (n/a)0.08 (n/a)0.02 (n/a)107.80 (n/a)76.80 (n/a)69.50 (n/a)67.10 (n/a)17.40 (n/a)

gelu_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.06 (-52.87%)0.05 (-56.10%)0.05 (-60.13%)0.03 (-42.37%)0.01 (-64.45%)265.60 (+73.48%)184.12 (+115.19%)173.80 (+150.79%)133.80 (+112.38%)48.73 (+28.44%)
1531781 — 2026-01-23 18:49:240.13 (n/a)0.11 (n/a)0.12 (n/a)0.05 (n/a)0.03 (n/a)153.10 (n/a)85.56 (n/a)69.30 (n/a)63.00 (n/a)37.94 (n/a)

gelu_4_cols_2_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.06 (-46.40%)0.06 (-47.57%)0.06 (-51.16%)0.04 (-39.83%)0.01 (-63.54%)183.60 (+66.15%)149.50 (+87.34%)144.20 (+104.54%)130.70 (+86.71%)20.19 (+15.51%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.11 (n/a)0.12 (n/a)0.07 (n/a)0.02 (n/a)110.50 (n/a)79.80 (n/a)70.50 (n/a)70.00 (n/a)17.48 (n/a)

gelu_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.06 (-49.30%)0.05 (-52.93%)0.05 (-60.02%)0.04 (-5.47%)0.01 (-75.28%)220.30 (+5.81%)178.92 (+81.20%)179.80 (+150.07%)138.50 (+97.29%)30.91 (-49.48%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.10 (n/a)0.11 (n/a)0.04 (n/a)0.03 (n/a)208.20 (n/a)98.74 (n/a)71.90 (n/a)70.20 (n/a)61.20 (n/a)

gelu_8_cols_2_channels_2048_tile_128

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.06 (-40.78%)0.04 (-54.91%)0.04 (-64.67%)0.03 (-49.86%)0.01 (-30.96%)266.10 (+99.48%)211.24 (+126.07%)232.90 (+182.99%)138.30 (+68.86%)51.05 (+127.22%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.09 (n/a)0.10 (n/a)0.06 (n/a)0.02 (n/a)133.40 (n/a)93.44 (n/a)82.30 (n/a)81.90 (n/a)22.47 (n/a)

gemm_1792x896x1152_64x32x48_8cols_ccolmaj

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
e40f553 — 2026-02-06 23:04:554.81 (-6.57%)4.35 (-11.15%)4.48 (-10.25%)3.84 (-10.08%)0.38 (+7.19%)2448.60 (+11.21%)2175.70 (+12.73%)2100.20 (+11.42%)1955.00 (+7.02%)195.34 (+26.81%)1892.23 (-6.57%)1711.06 (-11.15%)1761.41 (-10.25%)1510.84 (-10.08%)150.04 (+7.19%)
1531781 — 2026-01-23 18:49:245.15 (n/a)4.90 (n/a)4.99 (n/a)4.27 (n/a)0.36 (n/a)2201.70 (n/a)1929.94 (n/a)1885.00 (n/a)1826.70 (n/a)154.05 (n/a)2025.19 (n/a)1925.76 (n/a)1962.51 (n/a)1680.20 (n/a)139.97 (n/a)

gemm_192x384x64_48x96x16_4cols

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
e40f553 — 2026-02-06 23:04:551.23 (-47.71%)1.11 (-40.26%)1.10 (-38.20%)0.95 (-28.19%)0.12 (-68.31%)233.70 (+39.27%)202.20 (+62.75%)201.20 (+61.74%)179.30 (+91.36%)23.20 (-17.49%)52.64 (-47.71%)47.16 (-40.26%)46.90 (-38.20%)40.39 (-28.19%)5.30 (-68.31%)
1531781 — 2026-01-23 18:49:242.36 (n/a)1.85 (n/a)1.78 (n/a)1.32 (n/a)0.39 (n/a)167.80 (n/a)124.24 (n/a)124.40 (n/a)93.70 (n/a)28.11 (n/a)100.67 (n/a)78.94 (n/a)75.88 (n/a)56.25 (n/a)16.73 (n/a)

gemm_192x384x64_48x96x16_4cols_bcolmaj_ccolmaj

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
e40f553 — 2026-02-06 23:04:551.54 (-30.45%)1.05 (-49.02%)1.08 (-51.10%)0.54 (-66.28%)0.36 (+35.41%)409.80 (+196.53%)236.88 (+116.92%)205.50 (+104.48%)143.50 (+43.79%)101.38 (+513.01%)65.75 (-30.45%)44.75 (-49.02%)45.92 (-51.10%)23.03 (-66.28%)15.25 (+35.41%)
1531781 — 2026-01-23 18:49:242.22 (n/a)2.06 (n/a)2.20 (n/a)1.60 (n/a)0.26 (n/a)138.20 (n/a)109.20 (n/a)100.50 (n/a)99.80 (n/a)16.54 (n/a)94.54 (n/a)87.77 (n/a)93.91 (n/a)68.30 (n/a)11.26 (n/a)

gemm_2048x2048x2048_64x64x32_8_cols_0_bcolmaj_0_ccolmaj_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)GFLOP/s (max)GFLOP/s (mean)GFLOP/s (median)GFLOP/s (min)GFLOP/s (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3a053d3 — 2025-11-26 01:12:163.72 (-0.11%)3.55 (-2.75%)3.56 (-2.75%)3.36 (-5.33%)0.12 (+76.10%)2725.23 (-0.11%)2602.52 (-2.75%)2612.11 (-2.75%)2459.89 (-5.33%)86.27 (+76.10%)6984.00 (+5.63%)6608.60 (+2.91%)6577.00 (+2.83%)6304.00 (+0.11%)221.87 (+87.15%)
9e98049 — 2025-11-25 20:50:553.72 (n/a)3.65 (n/a)3.66 (n/a)3.54 (n/a)0.07 (n/a)2728.26 (n/a)2676.06 (n/a)2686.03 (n/a)2598.29 (n/a)48.99 (n/a)6612.00 (n/a)6422.00 (n/a)6396.00 (n/a)6297.00 (n/a)118.55 (n/a)

gemm_2048x2048x2048_64x64x32_8_cols_0_bcolmaj_1_ccolmaj_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)GFLOP/s (max)GFLOP/s (mean)GFLOP/s (median)GFLOP/s (min)GFLOP/s (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3a053d3 — 2025-11-26 01:12:163.40 (-3.14%)3.38 (-0.82%)3.38 (-0.82%)3.35 (+4.93%)0.02 (-84.51%)2493.81 (-3.14%)2475.27 (-0.82%)2479.06 (-0.82%)2453.92 (+4.93%)13.33 (-84.51%)7001.00 (-4.70%)6940.80 (+0.71%)6930.00 (+0.83%)6889.00 (+3.24%)37.43 (-84.76%)
9e98049 — 2025-11-25 20:50:553.51 (n/a)3.40 (n/a)3.41 (n/a)3.19 (n/a)0.12 (n/a)2574.53 (n/a)2495.72 (n/a)2499.62 (n/a)2338.67 (n/a)86.02 (n/a)7346.00 (n/a)6892.20 (n/a)6873.00 (n/a)6673.00 (n/a)245.57 (n/a)

gemm_2048x2048x2048_64x64x32_8_cols_1_bcolmaj_0_ccolmaj_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)GFLOP/s (max)GFLOP/s (mean)GFLOP/s (median)GFLOP/s (min)GFLOP/s (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3a053d3 — 2025-11-26 01:12:163.02 (+0.52%)2.87 (-2.05%)2.86 (-2.24%)2.70 (-6.16%)0.11 (+131.66%)2216.19 (+0.52%)2104.44 (-2.05%)2094.08 (-2.24%)1980.84 (-6.16%)80.25 (+131.67%)8673.00 (+6.56%)8175.60 (+2.21%)8204.00 (+2.29%)7752.00 (-0.51%)314.03 (+145.36%)
9e98049 — 2025-11-25 20:50:553.01 (n/a)2.93 (n/a)2.92 (n/a)2.88 (n/a)0.05 (n/a)2204.81 (n/a)2148.41 (n/a)2142.13 (n/a)2110.81 (n/a)34.64 (n/a)8139.00 (n/a)7998.60 (n/a)8020.00 (n/a)7792.00 (n/a)127.99 (n/a)

gemm_2048x2048x2048_64x64x64_1cols

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
e40f553 — 2026-02-06 23:04:550.52 (+0.29%)0.52 (-0.21%)0.52 (-0.33%)0.52 (-0.34%)0.00 (+414.59%)48578.30 (+0.34%)48472.28 (+0.21%)48520.60 (+0.33%)48205.90 (-0.29%)151.17 (+414.26%)356.39 (+0.29%)354.43 (-0.21%)354.07 (-0.33%)353.65 (-0.34%)1.11 (+414.56%)
1531781 — 2026-01-23 18:49:240.52 (n/a)0.52 (n/a)0.52 (n/a)0.52 (n/a)0.00 (n/a)48415.10 (n/a)48370.52 (n/a)48362.20 (n/a)48344.40 (n/a)29.40 (n/a)355.36 (n/a)355.17 (n/a)355.23 (n/a)354.85 (n/a)0.22 (n/a)

gemm_2048x2048x2048_64x64x64_2_cols_0_bcolmaj_0_ccolmaj_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:491.04 (+0.94%)1.03 (+0.23%)1.03 (+0.09%)1.02 (+0.07%)0.00 (+197.51%)24576.00 (-0.07%)24485.82 (-0.23%)24530.40 (-0.09%)24266.40 (-0.94%)111.26 (+194.35%)
0a6c11c — 2025-12-04 00:39:101.03 (n/a)1.03 (n/a)1.03 (n/a)1.02 (n/a)0.00 (n/a)24594.30 (n/a)24542.14 (n/a)24551.90 (n/a)24495.50 (n/a)37.80 (n/a)

gemm_2048x2048x2048_64x64x64_2_cols_0_bcolmaj_0_ccolmaj_0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
a4b6ffe — 2026-01-05 19:24:051.04 (+0.28%)1.03 (-0.15%)1.03 (+0.05%)1.02 (-0.52%)0.01 (+42.72%)24719.80 (+0.52%)24512.42 (+0.16%)24529.20 (-0.05%)24092.60 (-0.28%)252.92 (+43.15%)713.08 (+0.28%)700.92 (-0.15%)700.38 (+0.05%)694.98 (-0.52%)7.30 (+42.71%)
50ad37c — 2025-12-29 20:35:531.04 (n/a)1.03 (n/a)1.03 (n/a)1.02 (n/a)0.01 (n/a)24592.30 (n/a)24473.70 (n/a)24540.90 (n/a)24160.80 (n/a)176.68 (n/a)711.06 (n/a)702.00 (n/a)700.05 (n/a)698.59 (n/a)5.12 (n/a)

gemm_2048x2048x2048_64x64x64_2_cols_0_bcolmaj_1_ccolmaj_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.99 (+1.69%)0.97 (+0.29%)0.97 (-0.04%)0.97 (-0.15%)0.01 (+434.92%)26025.30 (+0.15%)25839.34 (-0.29%)25916.70 (+0.04%)25446.00 (-1.67%)202.05 (+425.44%)
0a6c11c — 2025-12-04 00:39:100.97 (n/a)0.97 (n/a)0.97 (n/a)0.97 (n/a)0.00 (n/a)25986.90 (n/a)25913.52 (n/a)25905.70 (n/a)25877.00 (n/a)38.45 (n/a)

gemm_2048x2048x2048_64x64x64_2_cols_0_bcolmaj_1_ccolmaj_0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
a4b6ffe — 2026-01-05 19:24:050.97 (-0.01%)0.97 (+0.03%)0.97 (+0.04%)0.97 (+0.00%)0.00 (-6.43%)25927.90 (-0.00%)25903.20 (-0.03%)25898.60 (-0.04%)25892.80 (+0.01%)14.05 (-6.56%)663.50 (-0.01%)663.23 (+0.03%)663.35 (+0.04%)662.60 (+0.00%)0.36 (-6.43%)
50ad37c — 2025-12-29 20:35:530.97 (n/a)0.97 (n/a)0.97 (n/a)0.97 (n/a)0.00 (n/a)25928.10 (n/a)25910.14 (n/a)25908.10 (n/a)25891.20 (n/a)15.04 (n/a)663.54 (n/a)663.06 (n/a)663.11 (n/a)662.60 (n/a)0.38 (n/a)

gemm_2048x2048x2048_64x64x64_2_cols_1_bcolmaj_0_ccolmaj_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.92 (-1.02%)0.92 (-0.66%)0.92 (-0.10%)0.91 (-0.78%)0.01 (-4.54%)27751.40 (+0.79%)27480.74 (+0.66%)27354.80 (+0.10%)27320.50 (+1.03%)182.69 (-2.84%)
0a6c11c — 2025-12-04 00:39:100.93 (n/a)0.92 (n/a)0.92 (n/a)0.91 (n/a)0.01 (n/a)27534.80 (n/a)27299.32 (n/a)27326.70 (n/a)27042.30 (n/a)188.04 (n/a)

gemm_2048x2048x2048_64x64x64_2_cols_1_bcolmaj_0_ccolmaj_0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
a4b6ffe — 2026-01-05 19:24:050.92 (+0.11%)0.91 (-0.12%)0.91 (-0.21%)0.91 (-0.24%)0.00 (+30.34%)27723.50 (+0.24%)27548.94 (+0.12%)27522.90 (+0.21%)27403.60 (-0.11%)124.26 (+30.58%)626.92 (+0.11%)623.62 (-0.12%)624.20 (-0.21%)619.69 (-0.24%)2.81 (+30.34%)
50ad37c — 2025-12-29 20:35:530.92 (n/a)0.91 (n/a)0.92 (n/a)0.91 (n/a)0.00 (n/a)27657.10 (n/a)27515.58 (n/a)27464.70 (n/a)27434.60 (n/a)95.16 (n/a)626.21 (n/a)624.37 (n/a)625.52 (n/a)621.17 (n/a)2.16 (n/a)

gemm_2048x2048x2048_64x64x64_2cols_bcolmaj

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
e40f553 — 2026-02-06 23:04:550.91 (-2.20%)0.90 (-1.03%)0.90 (-1.15%)0.90 (-0.22%)0.00 (-74.25%)27888.50 (+0.22%)27817.50 (+1.03%)27839.20 (+1.16%)27715.20 (+2.25%)71.95 (-73.58%)619.87 (-2.20%)617.60 (-1.03%)617.11 (-1.15%)616.02 (-0.22%)1.60 (-74.25%)
1531781 — 2026-01-23 18:49:240.93 (n/a)0.91 (n/a)0.91 (n/a)0.90 (n/a)0.01 (n/a)27826.70 (n/a)27533.84 (n/a)27519.90 (n/a)27105.20 (n/a)272.35 (n/a)633.82 (n/a)624.00 (n/a)624.27 (n/a)617.39 (n/a)6.21 (n/a)

gemm_2048x2048x2048_64x64x64_8_cols_0_bcolmaj_0_ccolmaj_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)GFLOP/s (max)GFLOP/s (mean)GFLOP/s (median)GFLOP/s (min)GFLOP/s (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
31ab24e — 2025-11-22 02:01:563.65 (-1.06%)3.59 (+0.83%)3.63 (-0.26%)3.45 (+2.66%)0.07 (-44.71%)2678.08 (-1.06%)2632.13 (+0.83%)2658.60 (-0.26%)2528.31 (+2.66%)53.86 (-44.71%)6795.00 (-2.59%)6529.80 (-0.92%)6462.00 (+0.26%)6415.00 (+1.07%)137.14 (-45.27%)
d9a64c5 — 2025-11-21 20:49:353.69 (n/a)3.56 (n/a)3.64 (n/a)3.36 (n/a)0.13 (n/a)2706.77 (n/a)2610.50 (n/a)2665.61 (n/a)2462.71 (n/a)97.41 (n/a)6976.00 (n/a)6590.40 (n/a)6445.00 (n/a)6347.00 (n/a)250.58 (n/a)

gemm_2048x2048x2048_64x64x64_8_cols_0_bcolmaj_0_ccolmaj_0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
d48746f — 2026-01-12 22:17:103.98 (+0.20%)3.84 (-2.95%)3.76 (-5.27%)3.72 (-4.97%)0.13 (+473.97%)6760.00 (+5.23%)6559.50 (+3.13%)6697.10 (+5.56%)6321.00 (-0.20%)218.68 (+499.40%)2717.91 (+0.20%)2621.44 (-2.95%)2565.27 (-5.27%)2541.40 (-4.97%)88.37 (+473.97%)
331dcca — 2026-01-08 18:34:103.97 (n/a)3.96 (n/a)3.97 (n/a)3.92 (n/a)0.02 (n/a)6423.80 (n/a)6360.32 (n/a)6344.30 (n/a)6333.90 (n/a)36.48 (n/a)2712.39 (n/a)2701.17 (n/a)2707.91 (n/a)2674.40 (n/a)15.40 (n/a)

gemm_2048x2048x2048_64x64x64_8_cols_0_bcolmaj_1_ccolmaj_0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
d48746f — 2026-01-12 22:17:103.75 (+0.28%)3.70 (+1.25%)3.71 (-0.21%)3.65 (+4.77%)0.04 (-66.90%)6901.80 (-4.56%)6795.46 (-1.30%)6790.20 (+0.21%)6715.90 (-0.28%)67.05 (-68.40%)2558.08 (+0.28%)2528.33 (+1.25%)2530.10 (-0.21%)2489.18 (+4.77%)24.81 (-66.90%)
331dcca — 2026-01-08 18:34:103.74 (n/a)3.66 (n/a)3.71 (n/a)3.48 (n/a)0.11 (n/a)7231.20 (n/a)6885.18 (n/a)6775.70 (n/a)6734.90 (n/a)212.21 (n/a)2550.87 (n/a)2497.04 (n/a)2535.50 (n/a)2375.78 (n/a)74.96 (n/a)

gemm_2048x2048x2048_64x64x64_8_cols_1_bcolmaj_0_ccolmaj_0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
d48746f — 2026-01-12 22:17:103.62 (-0.77%)3.54 (+0.14%)3.60 (-0.04%)3.44 (+2.17%)0.09 (-25.77%)7307.40 (-2.13%)7103.46 (-0.18%)6982.50 (+0.04%)6948.20 (+0.77%)182.95 (-26.55%)2472.55 (-0.77%)2419.80 (+0.14%)2460.41 (-0.04%)2351.04 (+2.17%)61.76 (-25.77%)
331dcca — 2026-01-08 18:34:103.65 (n/a)3.54 (n/a)3.61 (n/a)3.37 (n/a)0.12 (n/a)7466.20 (n/a)7116.40 (n/a)6979.60 (n/a)6895.10 (n/a)249.09 (n/a)2491.61 (n/a)2416.45 (n/a)2461.45 (n/a)2301.00 (n/a)83.20 (n/a)

gemm_2048x2048x2048_64x64x64_8cols_bcolmaj_ccolmaj

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
e40f553 — 2026-02-06 23:04:553.32 (-4.22%)3.21 (-6.09%)3.18 (-7.74%)3.15 (-4.11%)0.07 (-11.15%)7979.40 (+4.29%)7846.12 (+6.48%)7913.10 (+8.39%)7584.10 (+4.41%)156.52 (-3.70%)2265.25 (-4.22%)2190.31 (-6.09%)2171.06 (-7.74%)2153.03 (-4.11%)44.56 (-11.15%)
1531781 — 2026-01-23 18:49:243.46 (n/a)3.42 (n/a)3.45 (n/a)3.29 (n/a)0.07 (n/a)7651.20 (n/a)7368.34 (n/a)7300.80 (n/a)7263.70 (n/a)162.53 (n/a)2365.17 (n/a)2332.46 (n/a)2353.14 (n/a)2245.39 (n/a)50.15 (n/a)

gemm_384x1536x1792_32x48x64_4cols_bcolmaj

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
e40f553 — 2026-02-06 23:04:554.02 (-7.72%)3.69 (-12.99%)3.63 (-16.25%)3.51 (-8.77%)0.20 (-9.31%)2296.20 (+9.61%)2190.02 (+14.92%)2223.30 (+19.40%)2007.00 (+8.37%)112.98 (+6.70%)1053.27 (-7.72%)967.41 (-12.99%)950.83 (-16.25%)920.62 (-8.77%)52.16 (-9.31%)
1531781 — 2026-01-23 18:49:244.35 (n/a)4.24 (n/a)4.33 (n/a)3.85 (n/a)0.22 (n/a)2094.90 (n/a)1905.64 (n/a)1862.00 (n/a)1852.00 (n/a)105.88 (n/a)1141.45 (n/a)1111.86 (n/a)1135.29 (n/a)1009.08 (n/a)57.51 (n/a)

gemm_896x1792x640_32x64x80_8cols_ccolmaj

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
e40f553 — 2026-02-06 23:04:556.53 (-3.90%)5.31 (-17.54%)4.74 (-29.29%)4.52 (-15.29%)0.96 (+55.04%)1472.20 (+18.05%)1285.16 (+23.25%)1404.30 (+41.43%)1018.00 (+4.06%)217.07 (+89.43%)2018.78 (-3.90%)1639.10 (-17.54%)1463.53 (-29.29%)1395.98 (-15.29%)295.87 (+55.04%)
1531781 — 2026-01-23 18:49:246.80 (n/a)6.43 (n/a)6.70 (n/a)5.33 (n/a)0.62 (n/a)1247.10 (n/a)1042.70 (n/a)992.90 (n/a)978.30 (n/a)114.59 (n/a)2100.72 (n/a)1987.76 (n/a)2069.85 (n/a)1647.93 (n/a)190.83 (n/a)

layer_norm_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.05 (-53.44%)0.04 (-48.99%)0.05 (-43.62%)0.03 (-55.60%)0.01 (-57.52%)269.70 (+125.31%)189.96 (+94.95%)178.30 (+77.41%)153.50 (+114.69%)46.94 (+107.06%)
1531781 — 2026-01-23 18:49:240.11 (n/a)0.09 (n/a)0.08 (n/a)0.07 (n/a)0.02 (n/a)119.70 (n/a)97.44 (n/a)100.50 (n/a)71.50 (n/a)22.67 (n/a)

layer_norm_1_cols_2_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.05 (-52.38%)0.05 (-47.25%)0.05 (-44.21%)0.04 (-36.95%)0.01 (-70.15%)219.10 (+58.65%)181.42 (+83.66%)174.00 (+79.20%)160.50 (+110.08%)24.79 (-0.36%)
1531781 — 2026-01-23 18:49:240.11 (n/a)0.09 (n/a)0.08 (n/a)0.06 (n/a)0.02 (n/a)138.10 (n/a)98.78 (n/a)97.10 (n/a)76.40 (n/a)24.88 (n/a)

layer_norm_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.05 (-57.95%)0.04 (-55.76%)0.04 (-56.97%)0.02 (-59.99%)0.01 (-52.91%)350.20 (+149.96%)226.30 (+130.03%)219.30 (+132.31%)161.90 (+137.74%)73.87 (+182.96%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.09 (n/a)0.09 (n/a)0.06 (n/a)0.02 (n/a)140.10 (n/a)98.38 (n/a)94.40 (n/a)68.10 (n/a)26.11 (n/a)

layer_norm_2_cols_2_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.06 (-51.20%)0.05 (-49.46%)0.04 (-52.97%)0.04 (-40.02%)0.01 (-53.57%)217.40 (+66.72%)179.32 (+95.13%)198.80 (+112.62%)127.10 (+105.00%)40.94 (+59.98%)
1531781 — 2026-01-23 18:49:240.13 (n/a)0.09 (n/a)0.09 (n/a)0.06 (n/a)0.03 (n/a)130.40 (n/a)91.90 (n/a)93.50 (n/a)62.00 (n/a)25.59 (n/a)

layer_norm_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.05 (-57.03%)0.04 (-54.92%)0.05 (-55.31%)0.04 (-39.23%)0.01 (-70.78%)232.00 (+64.54%)195.22 (+113.26%)174.30 (+123.75%)170.00 (+132.88%)31.17 (+9.64%)
1531781 — 2026-01-23 18:49:240.11 (n/a)0.09 (n/a)0.11 (n/a)0.06 (n/a)0.02 (n/a)141.00 (n/a)91.54 (n/a)77.90 (n/a)73.00 (n/a)28.43 (n/a)

layer_norm_4_cols_2_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.05 (-54.65%)0.04 (-55.53%)0.04 (-59.45%)0.03 (-45.20%)0.01 (-67.20%)270.80 (+82.48%)222.74 (+115.96%)216.50 (+146.58%)162.60 (+120.62%)42.40 (+31.29%)
1531781 — 2026-01-23 18:49:240.11 (n/a)0.09 (n/a)0.09 (n/a)0.06 (n/a)0.02 (n/a)148.40 (n/a)103.14 (n/a)87.80 (n/a)73.70 (n/a)32.30 (n/a)

layer_norm_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.04 (-64.32%)0.04 (-63.90%)0.04 (-66.67%)0.02 (-53.59%)0.01 (-73.42%)331.60 (+115.46%)237.82 (+163.19%)232.20 (+200.00%)188.60 (+180.24%)55.75 (+55.44%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.10 (n/a)0.11 (n/a)0.05 (n/a)0.03 (n/a)153.90 (n/a)90.36 (n/a)77.40 (n/a)67.30 (n/a)35.86 (n/a)

layer_norm_8_cols_2_channels_2048_tile_128

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.04 (-57.88%)0.03 (-56.43%)0.04 (-56.75%)0.03 (-53.62%)0.00 (-66.28%)302.60 (+115.68%)242.14 (+127.36%)225.80 (+131.35%)219.30 (+137.34%)34.50 (+73.91%)
1531781 — 2026-01-23 18:49:240.09 (n/a)0.08 (n/a)0.08 (n/a)0.06 (n/a)0.01 (n/a)140.30 (n/a)106.50 (n/a)97.60 (n/a)92.40 (n/a)19.84 (n/a)

matrix_vector_mul_128x128_32_1col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:490.37 (+17.13%)0.30 (+24.87%)0.30 (+38.03%)0.23 (+12.47%)0.06 (+35.23%)
0a6c11c — 2025-12-04 00:39:100.32 (n/a)0.24 (n/a)0.22 (n/a)0.20 (n/a)0.04 (n/a)

matrix_vector_mul_128x128_32_1col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:100.26 (-1.81%)0.24 (+4.06%)0.24 (+0.52%)0.21 (+16.97%)0.02 (-48.16%)0.26 (-1.81%)0.24 (+4.06%)0.24 (+0.52%)0.20 (+16.97%)0.02 (-48.16%)
a4b6ffe — 2026-01-05 19:24:050.26 (n/a)0.23 (n/a)0.24 (n/a)0.18 (n/a)0.04 (n/a)0.26 (n/a)0.23 (n/a)0.24 (n/a)0.17 (n/a)0.04 (n/a)

matrix_vector_mul_128x128_32tsi_128tso_1col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
e40f553 — 2026-02-06 23:04:550.23 (-49.27%)0.21 (-40.31%)0.22 (-37.78%)0.16 (-26.09%)0.03 (-67.90%)0.22 (-49.27%)0.20 (-40.31%)0.21 (-37.78%)0.16 (-26.09%)0.03 (-67.90%)
1531781 — 2026-01-23 18:49:240.45 (n/a)0.35 (n/a)0.35 (n/a)0.22 (n/a)0.08 (n/a)0.44 (n/a)0.34 (n/a)0.34 (n/a)0.22 (n/a)0.08 (n/a)

matrix_vector_mul_2048x8192_1_1col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:4913.57 (+3.47%)13.15 (+1.09%)13.10 (+0.16%)12.52 (-1.58%)0.37 (+153.19%)
0a6c11c — 2025-12-04 00:39:1013.11 (n/a)13.01 (n/a)13.07 (n/a)12.72 (n/a)0.15 (n/a)

matrix_vector_mul_2048x8192_1_1col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:1013.43 (-0.07%)13.30 (+4.62%)13.27 (+1.29%)13.16 (+16.70%)0.12 (-86.69%)13.42 (-0.07%)13.29 (+4.62%)13.26 (+1.29%)13.16 (+16.70%)0.12 (-86.69%)
a4b6ffe — 2026-01-05 19:24:0513.44 (n/a)12.71 (n/a)13.10 (n/a)11.28 (n/a)0.87 (n/a)13.43 (n/a)12.70 (n/a)13.09 (n/a)11.27 (n/a)0.87 (n/a)

matrix_vector_mul_2048x8192_1_2col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:4925.06 (+2.47%)23.71 (+1.81%)24.13 (+0.50%)20.90 (+5.41%)1.48 (-15.60%)
0a6c11c — 2025-12-04 00:39:1024.46 (n/a)23.29 (n/a)24.01 (n/a)19.83 (n/a)1.75 (n/a)

matrix_vector_mul_2048x8192_1_2col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:1024.46 (-0.60%)21.07 (-8.03%)19.75 (-17.83%)18.08 (+1.18%)3.03 (+7.07%)24.44 (-0.60%)21.06 (-8.03%)19.74 (-17.83%)18.07 (+1.18%)3.03 (+7.07%)
a4b6ffe — 2026-01-05 19:24:0524.60 (n/a)22.91 (n/a)24.04 (n/a)17.87 (n/a)2.83 (n/a)24.59 (n/a)22.90 (n/a)24.02 (n/a)17.86 (n/a)2.83 (n/a)

matrix_vector_mul_2048x8192_1_4col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:4942.05 (+1.54%)40.50 (+12.60%)40.49 (+2.36%)38.98 (+97.53%)0.97 (-88.09%)
0a6c11c — 2025-12-04 00:39:1041.41 (n/a)35.97 (n/a)39.55 (n/a)19.73 (n/a)8.15 (n/a)

matrix_vector_mul_2048x8192_1_4col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:1042.58 (+1.51%)41.37 (+14.29%)41.25 (+4.89%)40.33 (+91.42%)0.92 (-89.18%)42.56 (+1.51%)41.34 (+14.29%)41.23 (+4.89%)40.30 (+91.42%)0.92 (-89.18%)
a4b6ffe — 2026-01-05 19:24:0541.95 (n/a)36.20 (n/a)39.33 (n/a)21.07 (n/a)8.54 (n/a)41.92 (n/a)36.17 (n/a)39.31 (n/a)21.05 (n/a)8.53 (n/a)

matrix_vector_mul_2048x8192_1_8col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:4946.62 (+3.11%)44.82 (+0.17%)44.78 (+0.17%)42.60 (-3.80%)1.60 (+367.72%)
0a6c11c — 2025-12-04 00:39:1045.21 (n/a)44.75 (n/a)44.70 (n/a)44.29 (n/a)0.34 (n/a)

matrix_vector_mul_2048x8192_1_8col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:1046.04 (+4.14%)45.06 (+3.47%)44.55 (+2.76%)44.30 (+2.84%)0.87 (+66.58%)46.01 (+4.14%)45.03 (+3.47%)44.52 (+2.76%)44.27 (+2.84%)0.87 (+66.58%)
a4b6ffe — 2026-01-05 19:24:0544.21 (n/a)43.55 (n/a)43.35 (n/a)43.07 (n/a)0.52 (n/a)44.18 (n/a)43.52 (n/a)43.32 (n/a)43.05 (n/a)0.52 (n/a)

matrix_vector_mul_2048x8192_1tsi_1024tso_2col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
e40f553 — 2026-02-06 23:04:5524.90 (+0.38%)24.50 (-0.36%)24.45 (-0.40%)24.23 (-0.70%)0.27 (+54.47%)24.88 (+0.38%)24.49 (-0.36%)24.43 (-0.40%)24.21 (-0.70%)0.27 (+54.47%)
1531781 — 2026-01-23 18:49:2424.80 (n/a)24.59 (n/a)24.55 (n/a)24.40 (n/a)0.18 (n/a)24.79 (n/a)24.58 (n/a)24.53 (n/a)24.38 (n/a)0.18 (n/a)

matrix_vector_mul_2048x8192_1tsi_2048tso_1col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
e40f553 — 2026-02-06 23:04:5513.36 (-1.28%)12.52 (-6.12%)13.17 (-2.01%)10.61 (-19.04%)1.19 (+476.78%)13.36 (-1.28%)12.52 (-6.12%)13.17 (-2.01%)10.60 (-19.04%)1.19 (+476.78%)
1531781 — 2026-01-23 18:49:2413.54 (n/a)13.34 (n/a)13.45 (n/a)13.10 (n/a)0.21 (n/a)13.53 (n/a)13.33 (n/a)13.44 (n/a)13.09 (n/a)0.21 (n/a)

matrix_vector_mul_2048x8192_1tsi_256tso_8col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
e40f553 — 2026-02-06 23:04:5543.56 (-6.95%)41.74 (-9.08%)42.23 (-8.95%)38.42 (-14.26%)1.95 (+118.53%)43.53 (-6.95%)41.72 (-9.08%)42.20 (-8.95%)38.40 (-14.26%)1.95 (+118.53%)
1531781 — 2026-01-23 18:49:2446.81 (n/a)45.91 (n/a)46.38 (n/a)44.81 (n/a)0.89 (n/a)46.79 (n/a)45.88 (n/a)46.35 (n/a)44.79 (n/a)0.89 (n/a)

matrix_vector_mul_2048x8192_1tsi_512tso_4col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
e40f553 — 2026-02-06 23:04:5544.30 (+6.77%)41.17 (+0.87%)40.35 (-1.88%)39.31 (-1.59%)1.93 (+196.76%)44.27 (+6.77%)41.14 (+0.87%)40.33 (-1.88%)39.29 (-1.59%)1.93 (+196.77%)
1531781 — 2026-01-23 18:49:2441.49 (n/a)40.81 (n/a)41.13 (n/a)39.95 (n/a)0.65 (n/a)41.46 (n/a)40.79 (n/a)41.10 (n/a)39.93 (n/a)0.65 (n/a)

matrix_vector_mul_8192x2048_4_1col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:4913.29 (+1.63%)13.19 (+6.08%)13.19 (+1.17%)13.09 (+26.86%)0.08 (-92.94%)
0a6c11c — 2025-12-04 00:39:1013.08 (n/a)12.44 (n/a)13.03 (n/a)10.32 (n/a)1.07 (n/a)

matrix_vector_mul_8192x2048_4_1col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:1013.68 (+3.42%)13.30 (+8.54%)13.38 (+7.25%)12.72 (+14.96%)0.35 (-66.09%)13.67 (+3.42%)13.29 (+8.54%)13.38 (+7.25%)12.71 (+14.96%)0.35 (-66.09%)
a4b6ffe — 2026-01-05 19:24:0513.23 (n/a)12.25 (n/a)12.48 (n/a)11.07 (n/a)1.04 (n/a)13.22 (n/a)12.24 (n/a)12.47 (n/a)11.06 (n/a)1.04 (n/a)

matrix_vector_mul_8192x2048_4_2col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:4925.58 (+4.24%)24.23 (+6.59%)24.03 (+0.29%)23.58 (+37.29%)0.70 (-74.97%)
0a6c11c — 2025-12-04 00:39:1024.54 (n/a)22.73 (n/a)23.96 (n/a)17.18 (n/a)2.79 (n/a)

matrix_vector_mul_8192x2048_4_2col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:1025.08 (+1.32%)24.68 (+13.42%)24.61 (+3.26%)24.47 (+48.32%)0.24 (-93.56%)25.07 (+1.32%)24.67 (+13.42%)24.59 (+3.26%)24.46 (+48.32%)0.24 (-93.56%)
a4b6ffe — 2026-01-05 19:24:0524.76 (n/a)21.76 (n/a)23.83 (n/a)16.50 (n/a)3.74 (n/a)24.74 (n/a)21.75 (n/a)23.82 (n/a)16.49 (n/a)3.73 (n/a)

matrix_vector_mul_8192x2048_4_4col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:4940.94 (-2.19%)39.74 (-1.10%)39.76 (-0.60%)38.17 (-2.14%)1.04 (+4.39%)
0a6c11c — 2025-12-04 00:39:1041.86 (n/a)40.18 (n/a)40.00 (n/a)39.00 (n/a)1.00 (n/a)

matrix_vector_mul_8192x2048_4_4col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:1040.74 (-1.09%)37.16 (-7.15%)40.25 (+1.47%)24.51 (-37.44%)7.07 (+736.13%)40.72 (-1.09%)37.13 (-7.15%)40.23 (+1.47%)24.50 (-37.44%)7.07 (+736.13%)
a4b6ffe — 2026-01-05 19:24:0541.19 (n/a)40.02 (n/a)39.67 (n/a)39.18 (n/a)0.85 (n/a)41.17 (n/a)39.99 (n/a)39.64 (n/a)39.16 (n/a)0.85 (n/a)

matrix_vector_mul_8192x2048_4_8col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:4946.71 (-1.04%)44.91 (-2.34%)44.71 (-2.98%)43.80 (-1.44%)0.96 (+6.93%)
0a6c11c — 2025-12-04 00:39:1047.20 (n/a)45.98 (n/a)46.08 (n/a)44.44 (n/a)0.90 (n/a)

matrix_vector_mul_8192x2048_4_8col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:1046.28 (+5.68%)44.89 (+14.59%)44.74 (+4.26%)44.20 (+62.80%)0.84 (-87.96%)46.25 (+5.68%)44.86 (+14.59%)44.71 (+4.26%)44.17 (+62.80%)0.84 (-87.96%)
a4b6ffe — 2026-01-05 19:24:0543.79 (n/a)39.17 (n/a)42.91 (n/a)27.15 (n/a)6.98 (n/a)43.77 (n/a)39.15 (n/a)42.88 (n/a)27.13 (n/a)6.98 (n/a)

matrix_vector_mul_8192x2048_4tsi_1024tso_1col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
e40f553 — 2026-02-06 23:04:5513.42 (-0.27%)12.96 (-2.71%)13.18 (-1.06%)12.15 (-7.94%)0.50 (+346.21%)13.42 (-0.27%)12.96 (-2.71%)13.17 (-1.06%)12.14 (-7.94%)0.50 (+346.21%)
1531781 — 2026-01-23 18:49:2413.46 (n/a)13.33 (n/a)13.32 (n/a)13.20 (n/a)0.11 (n/a)13.45 (n/a)13.32 (n/a)13.31 (n/a)13.19 (n/a)0.11 (n/a)

matrix_vector_mul_8192x2048_4tsi_1024tso_2col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
e40f553 — 2026-02-06 23:04:5524.90 (-1.44%)24.42 (-1.33%)24.26 (-1.41%)23.86 (-2.78%)0.46 (+55.86%)24.89 (-1.44%)24.41 (-1.33%)24.25 (-1.41%)23.85 (-2.78%)0.46 (+55.86%)
1531781 — 2026-01-23 18:49:2425.27 (n/a)24.75 (n/a)24.61 (n/a)24.55 (n/a)0.30 (n/a)25.25 (n/a)24.74 (n/a)24.59 (n/a)24.53 (n/a)0.30 (n/a)

matrix_vector_mul_8192x2048_4tsi_1024tso_4col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
e40f553 — 2026-02-06 23:04:5542.83 (+4.82%)41.32 (+3.28%)41.15 (+1.78%)39.89 (+2.65%)1.44 (+63.02%)42.80 (+4.82%)41.30 (+3.28%)41.13 (+1.78%)39.87 (+2.65%)1.44 (+63.02%)
1531781 — 2026-01-23 18:49:2440.86 (n/a)40.01 (n/a)40.43 (n/a)38.86 (n/a)0.89 (n/a)40.84 (n/a)39.98 (n/a)40.41 (n/a)38.84 (n/a)0.89 (n/a)

matrix_vector_mul_8192x2048_4tsi_1024tso_8col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
e40f553 — 2026-02-06 23:04:5545.93 (-0.33%)43.47 (-3.63%)42.76 (-5.99%)42.01 (-4.28%)1.70 (+76.02%)45.90 (-0.33%)43.44 (-3.63%)42.74 (-5.99%)41.99 (-4.28%)1.70 (+76.02%)
1531781 — 2026-01-23 18:49:2446.08 (n/a)45.11 (n/a)45.49 (n/a)43.89 (n/a)0.97 (n/a)46.05 (n/a)45.08 (n/a)45.46 (n/a)43.87 (n/a)0.97 (n/a)

mem_copy_16_cores_2_chans_2048_tile_128_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.07 (+22.03%)0.06 (+27.86%)0.06 (+35.49%)0.05 (+15.00%)0.01 (+49.69%)173.30 (-13.00%)137.94 (-21.32%)133.60 (-26.19%)123.40 (-18.06%)18.40 (+8.53%)
0a6c11c — 2025-12-04 00:39:100.05 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.00 (n/a)199.20 (n/a)175.32 (n/a)181.00 (n/a)150.60 (n/a)16.96 (n/a)

mem_copy_16_cores_2_chans_2048_tile_128_False0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.05 (-47.95%)0.04 (-51.49%)0.04 (-61.92%)0.03 (-18.81%)0.01 (-71.17%)240.60 (+23.20%)212.10 (+90.02%)227.50 (+162.70%)162.60 (+92.20%)32.26 (-32.30%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.08 (n/a)0.09 (n/a)0.04 (n/a)0.02 (n/a)195.30 (n/a)111.62 (n/a)86.60 (n/a)84.60 (n/a)47.65 (n/a)

mem_copy_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
003b49d — 2025-11-14 19:30:540.05 (-16.99%)0.05 (-12.47%)0.05 (-7.32%)0.04 (-11.17%)0.00 (-29.61%)197.00 (+12.57%)171.00 (+13.70%)164.00 (+7.89%)153.00 (+20.47%)17.49 (-4.32%)
0999a9a — 2025-11-13 19:24:080.06 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)175.00 (n/a)150.40 (n/a)152.00 (n/a)127.00 (n/a)18.28 (n/a)

mem_copy_1_cols_2_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
003b49d — 2025-11-14 19:30:540.06 (+5.56%)0.05 (-1.59%)0.05 (-1.39%)0.03 (-14.07%)0.01 (+25.03%)263.00 (+16.37%)166.00 (+4.53%)144.00 (+1.41%)126.00 (-5.26%)49.38 (+43.17%)
0999a9a — 2025-11-13 19:24:080.06 (n/a)0.05 (n/a)0.05 (n/a)0.03 (n/a)0.01 (n/a)226.00 (n/a)158.80 (n/a)142.00 (n/a)133.00 (n/a)34.49 (n/a)

mem_copy_1_cores_1_chans_2048_tile_2048_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.08 (+10.93%)0.07 (+7.79%)0.07 (+3.55%)0.05 (-1.73%)0.01 (+36.73%)150.20 (+1.76%)121.68 (-6.36%)125.60 (-3.38%)97.90 (-9.85%)18.56 (+21.98%)
0a6c11c — 2025-12-04 00:39:100.08 (n/a)0.06 (n/a)0.06 (n/a)0.06 (n/a)0.01 (n/a)147.60 (n/a)129.94 (n/a)130.00 (n/a)108.60 (n/a)15.21 (n/a)

mem_copy_1_cores_1_chans_2048_tile_2048_False0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.06 (-40.63%)0.05 (-46.01%)0.05 (-50.26%)0.04 (-48.02%)0.01 (-44.75%)227.20 (+92.38%)178.74 (+85.18%)169.80 (+101.18%)138.80 (+68.45%)33.07 (+80.82%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.09 (n/a)0.10 (n/a)0.07 (n/a)0.02 (n/a)118.10 (n/a)96.52 (n/a)84.40 (n/a)82.40 (n/a)18.29 (n/a)

mem_copy_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
003b49d — 2025-11-14 19:30:540.06 (-15.32%)0.05 (-17.09%)0.05 (-23.81%)0.04 (-15.54%)0.01 (-18.18%)193.00 (+18.40%)160.80 (+20.36%)168.00 (+31.25%)124.00 (+18.10%)22.92 (+11.39%)
0999a9a — 2025-11-13 19:24:080.07 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)163.00 (n/a)133.60 (n/a)128.00 (n/a)105.00 (n/a)20.58 (n/a)

mem_copy_2_cols_2_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
003b49d — 2025-11-14 19:30:540.06 (-23.02%)0.05 (-6.98%)0.05 (-2.13%)0.04 (-15.31%)0.01 (-33.29%)196.00 (+18.07%)147.00 (+6.52%)141.00 (+2.17%)126.00 (+29.90%)25.27 (+8.13%)
0999a9a — 2025-11-13 19:24:080.08 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)166.00 (n/a)138.00 (n/a)138.00 (n/a)97.00 (n/a)23.37 (n/a)

mem_copy_2_cores_1_chans_2048_tile_1024_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.10 (+31.37%)0.08 (+29.12%)0.08 (+29.06%)0.07 (+41.80%)0.01 (+0.23%)109.90 (-29.46%)98.30 (-23.38%)100.80 (-22.52%)79.40 (-23.87%)10.54 (-46.17%)
0a6c11c — 2025-12-04 00:39:100.08 (n/a)0.07 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)155.80 (n/a)128.30 (n/a)130.10 (n/a)104.30 (n/a)19.58 (n/a)

mem_copy_2_cores_1_chans_2048_tile_1024_False0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.06 (-51.93%)0.04 (-52.54%)0.04 (-57.63%)0.03 (-50.17%)0.01 (-61.45%)235.10 (+100.77%)190.50 (+106.62%)205.00 (+135.90%)146.40 (+107.95%)36.67 (+57.29%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.09 (n/a)0.09 (n/a)0.07 (n/a)0.02 (n/a)117.10 (n/a)92.20 (n/a)86.90 (n/a)70.40 (n/a)23.32 (n/a)

mem_copy_2_cores_2_chans_2048_tile_1024_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.10 (+25.02%)0.07 (+10.01%)0.07 (+7.53%)0.05 (-1.99%)0.01 (+77.08%)150.10 (+2.04%)114.60 (-7.22%)111.70 (-6.99%)83.90 (-20.02%)21.23 (+43.54%)
0a6c11c — 2025-12-04 00:39:100.08 (n/a)0.07 (n/a)0.07 (n/a)0.06 (n/a)0.01 (n/a)147.10 (n/a)123.52 (n/a)120.10 (n/a)104.90 (n/a)14.79 (n/a)

mem_copy_2_cores_2_chans_2048_tile_1024_False0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.06 (-34.97%)0.05 (-43.59%)0.05 (-47.11%)0.04 (-37.92%)0.01 (-39.75%)225.80 (+61.06%)179.38 (+76.45%)174.00 (+88.93%)127.60 (+53.92%)35.89 (+48.15%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.08 (n/a)0.09 (n/a)0.06 (n/a)0.02 (n/a)140.20 (n/a)101.66 (n/a)92.10 (n/a)82.90 (n/a)24.23 (n/a)

mem_copy_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
003b49d — 2025-11-14 19:30:540.07 (+11.97%)0.05 (+3.09%)0.05 (+2.10%)0.03 (-11.35%)0.01 (+41.65%)229.00 (+12.81%)156.00 (-0.51%)143.00 (-2.05%)117.00 (-10.69%)38.21 (+48.71%)
0999a9a — 2025-11-13 19:24:080.06 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)203.00 (n/a)156.80 (n/a)146.00 (n/a)131.00 (n/a)25.69 (n/a)

mem_copy_4_cols_2_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
003b49d — 2025-11-14 19:30:540.06 (+9.02%)0.05 (+5.59%)0.05 (+0.00%)0.04 (+3.63%)0.01 (+25.48%)193.00 (-3.50%)147.40 (-4.66%)147.00 (+0.00%)122.00 (-8.27%)25.24 (+7.48%)
0999a9a — 2025-11-13 19:24:080.06 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)200.00 (n/a)154.60 (n/a)147.00 (n/a)133.00 (n/a)23.48 (n/a)

mem_copy_4_cores_1_chans_2048_tile_512_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.09 (-3.98%)0.07 (+4.00%)0.07 (+7.04%)0.06 (+3.67%)0.01 (-15.76%)129.70 (-3.57%)113.30 (-4.50%)116.90 (-6.55%)90.40 (+4.15%)15.04 (-13.80%)
0a6c11c — 2025-12-04 00:39:100.09 (n/a)0.07 (n/a)0.07 (n/a)0.06 (n/a)0.01 (n/a)134.50 (n/a)118.64 (n/a)125.10 (n/a)86.80 (n/a)17.45 (n/a)

mem_copy_4_cores_1_chans_2048_tile_512_False0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.06 (-50.47%)0.05 (-51.08%)0.05 (-56.18%)0.04 (-38.69%)0.01 (-64.37%)209.30 (+63.13%)166.66 (+97.09%)163.30 (+128.07%)132.60 (+101.83%)31.15 (+17.92%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.10 (n/a)0.11 (n/a)0.06 (n/a)0.03 (n/a)128.30 (n/a)84.56 (n/a)71.60 (n/a)65.70 (n/a)26.42 (n/a)

mem_copy_4_cores_2_chans_2048_tile_512_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.08 (+9.57%)0.07 (+16.85%)0.07 (+19.58%)0.06 (+24.52%)0.01 (+1.05%)134.50 (-19.70%)116.64 (-14.92%)116.80 (-16.39%)98.00 (-8.75%)15.71 (-25.13%)
0a6c11c — 2025-12-04 00:39:100.08 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)167.50 (n/a)137.10 (n/a)139.70 (n/a)107.40 (n/a)20.99 (n/a)

mem_copy_4_cores_2_chans_2048_tile_512_False0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.05 (-52.75%)0.04 (-54.79%)0.04 (-57.98%)0.04 (-42.30%)0.00 (-75.31%)230.80 (+73.27%)199.98 (+115.08%)198.40 (+137.89%)173.80 (+111.69%)20.29 (-9.77%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.09 (n/a)0.10 (n/a)0.06 (n/a)0.02 (n/a)133.20 (n/a)92.98 (n/a)83.40 (n/a)82.10 (n/a)22.49 (n/a)

mem_copy_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
003b49d — 2025-11-14 19:30:540.05 (-13.48%)0.05 (-17.79%)0.05 (-15.33%)0.03 (-25.09%)0.01 (+6.67%)267.00 (+33.50%)176.80 (+24.51%)150.00 (+18.11%)141.00 (+15.57%)47.41 (+61.41%)
0999a9a — 2025-11-13 19:24:080.06 (n/a)0.06 (n/a)0.06 (n/a)0.04 (n/a)0.01 (n/a)200.00 (n/a)142.00 (n/a)127.00 (n/a)122.00 (n/a)29.37 (n/a)

mem_copy_8_cols_2_channels_2048_tile_128

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
003b49d — 2025-11-14 19:30:540.05 (+13.38%)0.05 (+23.85%)0.05 (+37.91%)0.05 (+30.72%)0.00 (-56.85%)166.00 (-23.50%)153.40 (-20.60%)153.00 (-27.49%)142.00 (-11.80%)7.63 (-70.60%)
0999a9a — 2025-11-13 19:24:080.05 (n/a)0.04 (n/a)0.04 (n/a)0.04 (n/a)0.01 (n/a)217.00 (n/a)193.20 (n/a)211.00 (n/a)161.00 (n/a)25.96 (n/a)

mem_copy_8_cores_1_chans_2048_tile_256_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.09 (+16.31%)0.08 (+15.34%)0.08 (+14.60%)0.07 (+19.30%)0.01 (-5.41%)117.30 (-16.15%)107.80 (-13.61%)108.50 (-12.78%)94.10 (-13.99%)8.29 (-32.01%)
0a6c11c — 2025-12-04 00:39:100.07 (n/a)0.07 (n/a)0.07 (n/a)0.06 (n/a)0.01 (n/a)139.90 (n/a)124.78 (n/a)124.40 (n/a)109.40 (n/a)12.20 (n/a)

mem_copy_8_cores_1_chans_2048_tile_256_False0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.06 (-35.81%)0.05 (-46.47%)0.04 (-53.91%)0.04 (-38.18%)0.01 (-34.60%)212.40 (+61.77%)186.66 (+87.07%)192.40 (+116.91%)135.40 (+55.81%)30.64 (+62.07%)
1531781 — 2026-01-23 18:49:240.09 (n/a)0.08 (n/a)0.09 (n/a)0.06 (n/a)0.01 (n/a)131.30 (n/a)99.78 (n/a)88.70 (n/a)86.90 (n/a)18.90 (n/a)

mem_copy_8_cores_2_chans_2048_tile_256_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.09 (+12.61%)0.07 (-2.24%)0.07 (-8.85%)0.06 (+5.13%)0.01 (+35.50%)136.50 (-4.88%)120.86 (+2.96%)125.10 (+9.64%)91.60 (-11.15%)15.72 (+10.46%)
0a6c11c — 2025-12-04 00:39:100.08 (n/a)0.07 (n/a)0.07 (n/a)0.06 (n/a)0.01 (n/a)143.50 (n/a)117.38 (n/a)114.10 (n/a)103.10 (n/a)14.23 (n/a)

mem_copy_8_cores_2_chans_2048_tile_256_False0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.07 (-28.60%)0.05 (-45.75%)0.04 (-52.97%)0.04 (-37.29%)0.01 (-17.34%)230.30 (+59.49%)182.84 (+87.14%)187.40 (+112.71%)112.10 (+40.12%)43.44 (+64.88%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.09 (n/a)0.09 (n/a)0.06 (n/a)0.02 (n/a)144.40 (n/a)97.70 (n/a)88.10 (n/a)80.00 (n/a)26.35 (n/a)

mha

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.18 (+0.16%)0.18 (+0.19%)0.18 (+0.12%)0.18 (+0.32%)0.00 (-52.24%)47253.90 (-0.32%)47213.70 (-0.19%)47210.90 (-0.12%)47167.70 (-0.16%)28.71 (-52.47%)
0a6c11c — 2025-12-04 00:39:100.18 (n/a)0.18 (n/a)0.18 (n/a)0.18 (n/a)0.00 (n/a)47405.00 (n/a)47301.72 (n/a)47265.90 (n/a)47241.80 (n/a)60.40 (n/a)

mha0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.18 (-0.87%)0.18 (-0.62%)0.18 (-0.57%)0.18 (-0.44%)0.00 (-59.32%)47578.10 (+0.44%)47526.30 (+0.62%)47503.50 (+0.57%)47484.60 (+0.87%)43.49 (-58.76%)
1531781 — 2026-01-23 18:49:240.18 (n/a)0.18 (n/a)0.18 (n/a)0.18 (n/a)0.00 (n/a)47367.50 (n/a)47232.68 (n/a)47232.00 (n/a)47073.00 (n/a)105.45 (n/a)

relu_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.07 (-41.99%)0.05 (-39.57%)0.06 (-44.06%)0.04 (-22.34%)0.01 (-66.35%)196.90 (+28.78%)157.60 (+55.09%)146.90 (+78.71%)125.90 (+72.47%)27.65 (-22.65%)
1531781 — 2026-01-23 18:49:240.11 (n/a)0.09 (n/a)0.10 (n/a)0.05 (n/a)0.03 (n/a)152.90 (n/a)101.62 (n/a)82.20 (n/a)73.00 (n/a)35.75 (n/a)

relu_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.06 (-40.76%)0.05 (-42.00%)0.05 (-39.09%)0.04 (-34.83%)0.01 (-39.03%)231.00 (+53.39%)173.92 (+71.76%)150.00 (+64.29%)140.20 (+68.92%)41.99 (+49.88%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.08 (n/a)0.09 (n/a)0.05 (n/a)0.02 (n/a)150.60 (n/a)101.26 (n/a)91.30 (n/a)83.00 (n/a)28.02 (n/a)

relu_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.05 (-47.99%)0.05 (-45.95%)0.05 (-49.54%)0.04 (-31.68%)0.00 (-74.24%)198.80 (+46.39%)173.02 (+79.85%)170.90 (+98.03%)159.70 (+92.18%)15.99 (-28.79%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.09 (n/a)0.09 (n/a)0.06 (n/a)0.02 (n/a)135.80 (n/a)96.20 (n/a)86.30 (n/a)83.10 (n/a)22.46 (n/a)

relu_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.06 (-48.89%)0.05 (-48.46%)0.05 (-53.77%)0.04 (-18.54%)0.00 (-81.78%)189.50 (+22.73%)168.88 (+80.97%)169.50 (+116.20%)148.30 (+95.65%)14.62 (-57.31%)
1531781 — 2026-01-23 18:49:240.11 (n/a)0.09 (n/a)0.10 (n/a)0.05 (n/a)0.02 (n/a)154.40 (n/a)93.32 (n/a)78.40 (n/a)75.80 (n/a)34.24 (n/a)

rms_norm_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.07 (-24.69%)0.05 (-38.15%)0.05 (-42.59%)0.04 (-38.09%)0.01 (-2.41%)212.40 (+61.52%)165.92 (+64.77%)160.90 (+74.13%)120.60 (+32.67%)35.94 (+107.21%)
1531781 — 2026-01-23 18:49:240.09 (n/a)0.08 (n/a)0.09 (n/a)0.06 (n/a)0.01 (n/a)131.50 (n/a)100.70 (n/a)92.40 (n/a)90.90 (n/a)17.35 (n/a)

rms_norm_1_cols_2_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.07 (-42.10%)0.05 (-51.01%)0.05 (-58.31%)0.04 (-27.75%)0.01 (-60.83%)201.00 (+38.43%)167.80 (+93.90%)172.90 (+139.81%)121.80 (+72.77%)29.14 (-11.16%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.10 (n/a)0.11 (n/a)0.06 (n/a)0.03 (n/a)145.20 (n/a)86.54 (n/a)72.10 (n/a)70.50 (n/a)32.80 (n/a)

rms_norm_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.06 (-33.92%)0.05 (-37.82%)0.05 (-39.21%)0.04 (-31.88%)0.01 (-44.62%)190.30 (+46.84%)156.94 (+59.78%)149.50 (+64.47%)135.90 (+51.34%)21.64 (+23.31%)
1531781 — 2026-01-23 18:49:240.09 (n/a)0.09 (n/a)0.09 (n/a)0.06 (n/a)0.01 (n/a)129.60 (n/a)98.22 (n/a)90.90 (n/a)89.80 (n/a)17.55 (n/a)

rms_norm_2_cols_2_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.07 (-44.20%)0.05 (-53.63%)0.05 (-52.45%)0.03 (-61.68%)0.01 (-6.42%)243.90 (+160.86%)168.50 (+126.72%)153.80 (+110.40%)111.80 (+79.17%)50.04 (+331.24%)
1531781 — 2026-01-23 18:49:240.13 (n/a)0.11 (n/a)0.11 (n/a)0.09 (n/a)0.02 (n/a)93.50 (n/a)74.32 (n/a)73.10 (n/a)62.40 (n/a)11.60 (n/a)

rms_norm_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.06 (-40.76%)0.04 (-48.53%)0.04 (-53.74%)0.04 (-41.60%)0.01 (-43.95%)223.90 (+71.31%)188.74 (+93.62%)195.00 (+116.19%)145.20 (+68.84%)28.67 (+53.64%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.09 (n/a)0.09 (n/a)0.06 (n/a)0.01 (n/a)130.70 (n/a)97.48 (n/a)90.20 (n/a)86.00 (n/a)18.66 (n/a)

rms_norm_4_cols_2_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.06 (-45.38%)0.05 (-52.91%)0.05 (-58.19%)0.04 (-36.57%)0.01 (-62.41%)197.90 (+57.69%)173.28 (+106.53%)175.70 (+139.05%)133.50 (+83.13%)24.44 (+5.08%)
1531781 — 2026-01-23 18:49:240.11 (n/a)0.10 (n/a)0.11 (n/a)0.07 (n/a)0.02 (n/a)125.50 (n/a)83.90 (n/a)73.50 (n/a)72.90 (n/a)23.26 (n/a)

rms_norm_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.07 (-24.86%)0.05 (-43.36%)0.04 (-51.73%)0.03 (-36.17%)0.01 (-18.32%)241.80 (+56.71%)190.52 (+78.62%)196.50 (+107.28%)125.20 (+33.19%)43.59 (+63.65%)
1531781 — 2026-01-23 18:49:240.09 (n/a)0.08 (n/a)0.09 (n/a)0.05 (n/a)0.01 (n/a)154.30 (n/a)106.66 (n/a)94.80 (n/a)94.00 (n/a)26.64 (n/a)

rms_norm_8_cols_2_channels_2048_tile_128

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.04 (-53.84%)0.04 (-51.28%)0.04 (-57.41%)0.04 (-26.12%)0.00 (-85.26%)227.00 (+35.36%)212.72 (+94.51%)215.10 (+134.83%)192.60 (+116.65%)14.44 (-56.91%)
1531781 — 2026-01-23 18:49:240.09 (n/a)0.08 (n/a)0.09 (n/a)0.05 (n/a)0.02 (n/a)167.70 (n/a)109.36 (n/a)91.60 (n/a)88.90 (n/a)33.52 (n/a)

rope_1_cols_2_channels_4096_tile_4096_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1b2d230 — 2026-01-13 22:38:030.19 (-21.66%)0.17 (-8.53%)0.18 (+2.09%)0.14 (-6.71%)0.02 (-38.07%)174.10 (+7.20%)150.06 (+8.11%)139.60 (-2.04%)129.80 (+27.63%)20.36 (-10.65%)
d48746f — 2026-01-12 22:17:100.24 (n/a)0.18 (n/a)0.17 (n/a)0.15 (n/a)0.04 (n/a)162.40 (n/a)138.80 (n/a)142.50 (n/a)101.70 (n/a)22.79 (n/a)

rope_1c_32rows_512cols_32arows_0m

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.73 (-33.38%)0.63 (-36.04%)0.61 (-42.00%)0.56 (-24.17%)0.08 (-48.53%)176.40 (+31.84%)157.08 (+54.70%)160.90 (+72.45%)133.80 (+50.17%)18.67 (+0.41%)
1531781 — 2026-01-23 18:49:241.10 (n/a)0.99 (n/a)1.05 (n/a)0.73 (n/a)0.15 (n/a)133.80 (n/a)101.54 (n/a)93.30 (n/a)89.10 (n/a)18.60 (n/a)

rope_1c_32rows_512cols_8arows_0m

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.66 (+18.23%)0.50 (-3.90%)0.48 (-9.01%)0.41 (-14.95%)0.10 (+231.15%)180.70 (+17.57%)150.52 (+6.87%)154.00 (+9.92%)112.10 (-15.40%)27.65 (+229.19%)
1531781 — 2026-01-23 18:49:240.56 (n/a)0.52 (n/a)0.53 (n/a)0.48 (n/a)0.03 (n/a)153.70 (n/a)140.84 (n/a)140.10 (n/a)132.50 (n/a)8.40 (n/a)

rope_2_cols_2_channels_4096_tile_2048_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1b2d230 — 2026-01-13 22:38:030.21 (-0.56%)0.17 (-10.62%)0.16 (-13.58%)0.13 (-13.33%)0.03 (+17.05%)183.70 (+15.39%)151.76 (+12.77%)155.20 (+15.73%)117.30 (+0.60%)24.24 (+35.73%)
d48746f — 2026-01-12 22:17:100.21 (n/a)0.19 (n/a)0.18 (n/a)0.15 (n/a)0.02 (n/a)159.20 (n/a)134.58 (n/a)134.10 (n/a)116.60 (n/a)17.86 (n/a)

rope_2c_32rows_512cols_32arows_0m

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.78 (-37.24%)0.64 (-33.76%)0.66 (-27.19%)0.47 (-25.70%)0.12 (-50.57%)211.20 (+34.52%)158.82 (+46.92%)148.80 (+37.40%)125.40 (+59.34%)33.72 (+8.45%)
1531781 — 2026-01-23 18:49:241.25 (n/a)0.97 (n/a)0.91 (n/a)0.63 (n/a)0.25 (n/a)157.00 (n/a)108.10 (n/a)108.30 (n/a)78.70 (n/a)31.09 (n/a)

rope_2c_32rows_512cols_8arows_0m

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.61 (-3.01%)0.53 (-6.24%)0.55 (-3.06%)0.41 (-15.90%)0.08 (+49.63%)179.10 (+18.92%)140.98 (+7.96%)134.50 (+3.14%)120.40 (+3.08%)23.15 (+84.24%)
1531781 — 2026-01-23 18:49:240.63 (n/a)0.57 (n/a)0.57 (n/a)0.49 (n/a)0.05 (n/a)150.60 (n/a)130.58 (n/a)130.40 (n/a)116.80 (n/a)12.57 (n/a)

rope_4_cols_2_channels_4096_tile_1024_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1b2d230 — 2026-01-13 22:38:030.22 (+18.92%)0.17 (+6.24%)0.17 (+5.57%)0.14 (-0.50%)0.03 (+90.77%)171.10 (+0.47%)147.40 (-4.39%)148.10 (-5.31%)110.10 (-15.89%)23.59 (+60.67%)
d48746f — 2026-01-12 22:17:100.19 (n/a)0.16 (n/a)0.16 (n/a)0.14 (n/a)0.02 (n/a)170.30 (n/a)154.16 (n/a)156.40 (n/a)130.90 (n/a)14.69 (n/a)

rope_8_cols_2_channels_4096_tile_512_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1b2d230 — 2026-01-13 22:38:030.17 (-1.30%)0.15 (+6.63%)0.15 (-8.78%)0.14 (+39.68%)0.01 (-67.35%)174.90 (-28.41%)163.42 (-10.36%)167.40 (+9.63%)146.50 (+1.31%)11.19 (-76.17%)
d48746f — 2026-01-12 22:17:100.17 (n/a)0.14 (n/a)0.16 (n/a)0.10 (n/a)0.03 (n/a)244.30 (n/a)182.30 (n/a)152.70 (n/a)144.60 (n/a)46.98 (n/a)

rope_8c_32rows_512cols_32arows_0m

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.65 (-51.18%)0.58 (-35.07%)0.61 (-27.80%)0.45 (-24.47%)0.08 (-69.20%)216.50 (+32.42%)172.44 (+46.91%)160.00 (+38.53%)150.70 (+104.76%)27.53 (-14.17%)
1531781 — 2026-01-23 18:49:241.34 (n/a)0.89 (n/a)0.85 (n/a)0.60 (n/a)0.27 (n/a)163.50 (n/a)117.38 (n/a)115.50 (n/a)73.60 (n/a)32.08 (n/a)

rope_8c_32rows_512cols_8arows_0m

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.60 (+11.00%)0.41 (-20.98%)0.37 (-29.47%)0.31 (-35.10%)0.11 (+331.33%)235.60 (+54.09%)190.32 (+32.79%)198.10 (+41.80%)122.10 (-9.89%)42.29 (+466.52%)
1531781 — 2026-01-23 18:49:240.54 (n/a)0.52 (n/a)0.53 (n/a)0.48 (n/a)0.03 (n/a)152.90 (n/a)143.32 (n/a)139.70 (n/a)135.50 (n/a)7.46 (n/a)

sigmoid_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.07 (-42.10%)0.06 (-41.31%)0.05 (-53.99%)0.05 (-6.32%)0.01 (-72.46%)163.60 (+6.79%)142.34 (+56.56%)149.00 (+117.20%)118.00 (+72.77%)19.19 (-47.85%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.10 (n/a)0.12 (n/a)0.05 (n/a)0.03 (n/a)153.20 (n/a)90.92 (n/a)68.60 (n/a)68.30 (n/a)36.80 (n/a)

sigmoid_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.07 (-40.09%)0.06 (-42.46%)0.06 (-50.74%)0.04 (-23.97%)0.01 (-60.24%)185.30 (+31.61%)148.20 (+64.96%)140.00 (+103.19%)113.70 (+66.96%)29.75 (-7.16%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.10 (n/a)0.12 (n/a)0.06 (n/a)0.03 (n/a)140.80 (n/a)89.84 (n/a)68.90 (n/a)68.10 (n/a)32.04 (n/a)

sigmoid_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.07 (-30.37%)0.06 (-30.44%)0.05 (-44.01%)0.05 (+2.10%)0.01 (-58.08%)165.90 (-2.07%)145.34 (+36.37%)156.20 (+78.72%)120.60 (+43.57%)21.14 (-41.87%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.08 (n/a)0.09 (n/a)0.05 (n/a)0.02 (n/a)169.40 (n/a)106.58 (n/a)87.40 (n/a)84.00 (n/a)36.37 (n/a)

sigmoid_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.07 (-45.45%)0.05 (-47.56%)0.05 (-59.52%)0.04 (-21.64%)0.01 (-55.93%)200.70 (+27.59%)161.46 (+79.92%)169.80 (+147.16%)120.90 (+83.46%)38.57 (-0.71%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.10 (n/a)0.12 (n/a)0.05 (n/a)0.03 (n/a)157.30 (n/a)89.74 (n/a)68.70 (n/a)65.90 (n/a)38.85 (n/a)

silu_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.07 (-29.73%)0.05 (-37.33%)0.05 (-45.50%)0.04 (-36.12%)0.01 (-41.52%)224.40 (+56.59%)164.72 (+57.06%)150.90 (+83.58%)114.70 (+42.31%)42.97 (+33.31%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.08 (n/a)0.10 (n/a)0.06 (n/a)0.02 (n/a)143.30 (n/a)104.88 (n/a)82.20 (n/a)80.60 (n/a)32.24 (n/a)

silu_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.06 (-40.40%)0.05 (-42.59%)0.05 (-52.04%)0.04 (-38.02%)0.01 (-59.87%)224.10 (+61.34%)174.14 (+68.90%)173.60 (+108.65%)136.90 (+67.77%)31.73 (+12.12%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.08 (n/a)0.10 (n/a)0.06 (n/a)0.02 (n/a)138.90 (n/a)103.10 (n/a)83.20 (n/a)81.60 (n/a)28.30 (n/a)

silu_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.07 (-41.47%)0.06 (-47.54%)0.05 (-58.02%)0.04 (-28.70%)0.01 (-51.78%)192.60 (+40.17%)153.50 (+84.14%)166.70 (+138.14%)116.30 (+70.78%)33.19 (+9.64%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.11 (n/a)0.12 (n/a)0.06 (n/a)0.03 (n/a)137.40 (n/a)83.36 (n/a)70.00 (n/a)68.10 (n/a)30.27 (n/a)

silu_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.05 (-53.98%)0.04 (-55.25%)0.04 (-58.13%)0.04 (-51.63%)0.01 (-53.77%)227.00 (+106.74%)199.18 (+123.30%)210.00 (+138.91%)154.50 (+117.30%)30.15 (+107.84%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.09 (n/a)0.09 (n/a)0.07 (n/a)0.02 (n/a)109.80 (n/a)89.20 (n/a)87.90 (n/a)71.10 (n/a)14.50 (n/a)

softmax_1_cols_2_channels_4096_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.21 (+20.19%)0.16 (+13.84%)0.16 (+8.19%)0.12 (+26.61%)0.04 (+17.18%)135.50 (-20.99%)104.68 (-12.65%)102.70 (-7.56%)76.50 (-16.85%)22.55 (-26.21%)
0f0da0b — 2026-01-14 16:31:000.18 (n/a)0.14 (n/a)0.15 (n/a)0.10 (n/a)0.03 (n/a)171.50 (n/a)119.84 (n/a)111.10 (n/a)92.00 (n/a)30.56 (n/a)

softmax_2_cols_2_channels_32768_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.99 (n/a)0.76 (n/a)0.77 (n/a)0.51 (n/a)0.17 (n/a)259.40 (n/a)180.34 (n/a)170.70 (n/a)132.80 (n/a)47.05 (n/a)

softmax_2_cols_2_channels_32768_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:551.02 (n/a)0.85 (n/a)0.90 (n/a)0.66 (n/a)0.16 (n/a)199.30 (n/a)158.50 (n/a)145.20 (n/a)128.30 (n/a)32.19 (n/a)

softmax_2_cols_2_channels_4096_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.22 (+40.26%)0.18 (+43.33%)0.20 (+64.08%)0.13 (+24.34%)0.04 (+68.98%)127.80 (-19.62%)94.74 (-28.95%)83.40 (-39.08%)73.90 (-28.67%)24.46 (-5.31%)
0f0da0b — 2026-01-14 16:31:000.16 (n/a)0.13 (n/a)0.12 (n/a)0.10 (n/a)0.03 (n/a)159.00 (n/a)133.34 (n/a)136.90 (n/a)103.60 (n/a)25.83 (n/a)

softmax_2_cols_2_channels_4096_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.24 (+85.21%)0.17 (+69.97%)0.15 (+54.16%)0.11 (+37.21%)0.05 (+203.82%)148.10 (-27.12%)102.78 (-37.88%)106.70 (-35.10%)69.60 (-45.96%)31.81 (+15.98%)
0f0da0b — 2026-01-14 16:31:000.13 (n/a)0.10 (n/a)0.10 (n/a)0.08 (n/a)0.02 (n/a)203.20 (n/a)165.46 (n/a)164.40 (n/a)128.80 (n/a)27.43 (n/a)

softmax_4_cols_4_channels_32768_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.77 (n/a)0.57 (n/a)0.65 (n/a)0.33 (n/a)0.19 (n/a)400.80 (n/a)257.22 (n/a)200.30 (n/a)170.00 (n/a)101.12 (n/a)

swiglu

No metrics available.

swiglu_decode_1x2048x2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.00 (-15.46%)0.00 (-27.20%)0.00 (-28.17%)0.00 (-33.90%)0.00 (+1834.83%)5187.20 (+51.28%)4720.84 (+38.38%)4750.90 (+39.22%)4015.10 (+18.29%)390.23 (+3298.45%)
0a6c11c — 2025-12-04 00:39:100.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)3428.80 (n/a)3411.62 (n/a)3412.60 (n/a)3394.20 (n/a)11.48 (n/a)

swiglu_decode_1x2048x2048_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:55n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)
1531781 — 2026-01-23 18:49:240.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)5111.40 (n/a)4324.88 (n/a)4736.10 (n/a)3410.70 (n/a)843.97 (n/a)

swiglu_prefill_256x2048x2048_0

No metrics available.

tanh_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.06 (-36.13%)0.05 (-34.47%)0.05 (-23.19%)0.03 (-42.98%)0.01 (-29.82%)247.00 (+75.43%)169.04 (+54.88%)150.90 (+30.20%)134.00 (+56.54%)47.23 (+100.84%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.08 (n/a)0.07 (n/a)0.06 (n/a)0.02 (n/a)140.80 (n/a)109.14 (n/a)115.90 (n/a)85.60 (n/a)23.52 (n/a)

tanh_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.07 (-32.31%)0.06 (-36.67%)0.07 (-24.86%)0.03 (-62.44%)0.02 (+83.84%)275.10 (+166.31%)154.10 (+75.07%)116.80 (+33.03%)114.90 (+47.69%)69.08 (+616.82%)
1531781 — 2026-01-23 18:49:240.11 (n/a)0.09 (n/a)0.09 (n/a)0.08 (n/a)0.01 (n/a)103.30 (n/a)88.02 (n/a)87.80 (n/a)77.80 (n/a)9.64 (n/a)

tanh_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.06 (-33.65%)0.05 (-45.29%)0.04 (-53.35%)0.04 (-41.49%)0.01 (-17.88%)207.40 (+70.84%)173.00 (+85.22%)185.00 (+114.37%)127.60 (+50.83%)32.87 (+107.99%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.09 (n/a)0.09 (n/a)0.07 (n/a)0.01 (n/a)121.40 (n/a)93.40 (n/a)86.30 (n/a)84.60 (n/a)15.80 (n/a)

tanh_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.07 (-36.32%)0.05 (-43.28%)0.05 (-53.82%)0.04 (-28.74%)0.01 (-50.66%)229.80 (+40.29%)172.88 (+68.01%)174.00 (+116.42%)112.50 (+56.90%)43.30 (+8.28%)
1531781 — 2026-01-23 18:49:240.11 (n/a)0.09 (n/a)0.10 (n/a)0.05 (n/a)0.03 (n/a)163.80 (n/a)102.90 (n/a)80.40 (n/a)71.70 (n/a)39.99 (n/a)

transpose_2048_M_64_N_1_cols_1_channels_64_m_64_n_8_s

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:493.93 (-7.02%)3.68 (+2.48%)3.66 (+1.82%)3.45 (+15.62%)0.15 (-63.64%)151.80 (-13.50%)142.70 (-3.59%)143.40 (-1.78%)133.50 (+7.49%)5.90 (-66.22%)
0a6c11c — 2025-12-04 00:39:104.22 (n/a)3.59 (n/a)3.59 (n/a)2.99 (n/a)0.42 (n/a)175.50 (n/a)148.02 (n/a)146.00 (n/a)124.20 (n/a)17.46 (n/a)

transpose_2048_M_64_N_1_cols_1_channels_64_m_64_n_8_s0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:553.73 (-24.53%)2.85 (-34.41%)2.72 (-37.85%)2.03 (-45.65%)0.71 (+41.71%)258.20 (+84.03%)193.46 (+58.60%)192.90 (+60.88%)140.40 (+32.45%)48.22 (+240.66%)
1531781 — 2026-01-23 18:49:244.95 (n/a)4.34 (n/a)4.37 (n/a)3.74 (n/a)0.50 (n/a)140.30 (n/a)121.98 (n/a)119.90 (n/a)106.00 (n/a)14.15 (n/a)

transpose_2048_M_64_N_1_cols_2_channels_64_m_64_n_8_s

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:495.11 (+15.57%)3.85 (-3.32%)3.40 (-18.64%)3.29 (+7.09%)0.71 (+49.14%)159.60 (-6.61%)140.18 (+4.80%)154.20 (+22.97%)102.60 (-13.49%)22.70 (+19.77%)
0a6c11c — 2025-12-04 00:39:104.42 (n/a)3.99 (n/a)4.18 (n/a)3.07 (n/a)0.48 (n/a)170.90 (n/a)133.76 (n/a)125.40 (n/a)118.60 (n/a)18.95 (n/a)

transpose_2048_M_64_N_1_cols_2_channels_64_m_64_n_8_s0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:553.91 (-35.58%)3.13 (-37.40%)3.29 (-35.40%)2.08 (-49.19%)0.76 (+2.81%)252.00 (+96.72%)177.18 (+65.81%)159.50 (+54.85%)134.00 (+55.27%)48.90 (+212.07%)
1531781 — 2026-01-23 18:49:246.07 (n/a)4.99 (n/a)5.09 (n/a)4.09 (n/a)0.74 (n/a)128.10 (n/a)106.86 (n/a)103.00 (n/a)86.30 (n/a)15.67 (n/a)

weighted_rms_norm_1_cols_2_channels_2048_weights_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.09 (-13.77%)0.08 (-12.46%)0.07 (-19.91%)0.06 (+9.21%)0.01 (-31.52%)212.90 (+37.35%)165.86 (+65.99%)167.80 (+87.28%)133.00 (+73.86%)31.75 (+1.24%)
1531781 — 2026-01-23 18:49:240.11 (n/a)0.09 (n/a)0.09 (n/a)0.05 (n/a)0.02 (n/a)155.00 (n/a)99.92 (n/a)89.60 (n/a)76.50 (n/a)31.36 (n/a)

weighted_rms_norm_2_cols_2_channels_2048_weights_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.08 (-15.73%)0.07 (-21.76%)0.07 (-29.51%)0.05 (-5.74%)0.02 (-24.79%)221.00 (+32.57%)160.38 (+56.13%)156.90 (+77.29%)123.30 (+48.38%)40.03 (+11.61%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.09 (n/a)0.09 (n/a)0.05 (n/a)0.02 (n/a)166.70 (n/a)102.72 (n/a)88.50 (n/a)83.10 (n/a)35.87 (n/a)

weighted_rms_norm_4_cols_2_channels_2048_weights_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.06 (-35.39%)0.05 (-40.13%)0.05 (-42.40%)0.04 (-35.52%)0.01 (-38.45%)207.70 (+74.54%)181.10 (+87.71%)175.50 (+95.22%)155.40 (+74.02%)21.45 (+67.46%)
1531781 — 2026-01-23 18:49:240.09 (n/a)0.09 (n/a)0.09 (n/a)0.07 (n/a)0.01 (n/a)119.00 (n/a)96.48 (n/a)89.90 (n/a)89.30 (n/a)12.81 (n/a)

weighted_rms_norm_8_cols_2_channels_2048_weights_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
e40f553 — 2026-02-06 23:04:550.05 (-55.29%)0.04 (-60.19%)0.04 (-60.37%)0.02 (-64.51%)0.01 (-48.95%)406.50 (+199.34%)247.62 (+175.38%)214.60 (+168.25%)176.70 (+137.82%)90.85 (+252.66%)
1531781 — 2026-01-23 18:49:240.11 (n/a)0.10 (n/a)0.10 (n/a)0.06 (n/a)0.02 (n/a)135.80 (n/a)89.92 (n/a)80.00 (n/a)74.30 (n/a)25.76 (n/a)

@github-actions
Copy link
Contributor

github-actions bot commented Feb 6, 2026

📊 Test Results for Test Example Applications

e40f553 (2026_02_06_23_15_29)

IRONCLAD

Tested on 2026_02_06_23_15_29 at commit e40f553.

Test Checks TTFT (mean)TPS (mean)
llama_3.2_1b_prompt_13_tokens_1 ❌ 0/1 n/a n/a
llama_3.2_1b_prompt_13_tokens_40 ❌ 0/1 n/a n/a
llama_3.2_1b_prompt_2048_tokens_1 ❌ 0/1 n/a n/a
llama_3.2_1b_prompt_2048_tokens_40 ❌ 0/1 n/a n/a
📈 Trends (vs main branch) for Test Example Applications

e40f553 (2026_02_06_23_15_29)

IRONCLAD Trends

llama_3.2_1b

Commit/Date Num Tokens (max)Num Tokens (mean)Num Tokens (median)Num Tokens (min)Num Tokens (stddev)TPS (max)TPS (mean)TPS (median)TPS (min)TPS (stddev)TTFT (max)TTFT (mean)TTFT (median)TTFT (min)TTFT (stddev)Total (max)Total (mean)Total (median)Total (min)Total (stddev)
130b6ea — 2025-12-05 21:33:1240.00 (+0.00%)40.00 (+0.00%)40.00 (+0.00%)40.00 (+0.00%)0.00 (n/a)4.71 (-0.42%)4.64 (-0.09%)4.64 (+0.65%)4.55 (-0.22%)0.05 (-17.66%)4.41 (-0.34%)4.39 (-0.19%)4.38 (-0.33%)4.37 (-0.15%)0.01 (-25.90%)12.96 (-0.00%)12.80 (+0.07%)12.80 (-0.23%)12.67 (+0.44%)0.09 (-21.12%)
0a6c11c — 2025-12-03 23:35:1540.00 (n/a)40.00 (n/a)40.00 (n/a)40.00 (n/a)0.00 (n/a)4.73 (n/a)4.64 (n/a)4.61 (n/a)4.56 (n/a)0.06 (n/a)4.42 (n/a)4.40 (n/a)4.40 (n/a)4.37 (n/a)0.02 (n/a)12.96 (n/a)12.79 (n/a)12.83 (n/a)12.62 (n/a)0.12 (n/a)

llama_3.2_1b_prompt_13_tokens_1

Commit/Date Num_Tokens (max)Num_Tokens (mean)Num_Tokens (median)Num_Tokens (min)Num_Tokens (stddev)TPS (max)TPS (mean)TPS (median)TPS (min)TPS (stddev)TTFT (max)TTFT (mean)TTFT (median)TTFT (min)TTFT (stddev)
e40f553 — 2026-02-06 23:15:19n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)
1531781 — 2026-01-23 18:40:341.00 (n/a)1.00 (n/a)1.00 (n/a)1.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.61 (n/a)0.61 (n/a)0.61 (n/a)0.61 (n/a)0.00 (n/a)

llama_3.2_1b_prompt_13_tokens_40

Commit/Date Num_Tokens (max)Num_Tokens (mean)Num_Tokens (median)Num_Tokens (min)Num_Tokens (stddev)TPS (max)TPS (mean)TPS (median)TPS (min)TPS (stddev)TTFT (max)TTFT (mean)TTFT (median)TTFT (min)TTFT (stddev)
e40f553 — 2026-02-06 23:15:19n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)
1531781 — 2026-01-23 18:40:3440.00 (n/a)40.00 (n/a)40.00 (n/a)40.00 (n/a)0.00 (n/a)4.34 (n/a)4.34 (n/a)4.34 (n/a)4.34 (n/a)0.00 (n/a)0.61 (n/a)0.61 (n/a)0.61 (n/a)0.61 (n/a)0.00 (n/a)

llama_3.2_1b_prompt_2048_tokens_1

Commit/Date Num_Tokens (max)Num_Tokens (mean)Num_Tokens (median)Num_Tokens (min)Num_Tokens (stddev)TPS (max)TPS (mean)TPS (median)TPS (min)TPS (stddev)TTFT (max)TTFT (mean)TTFT (median)TTFT (min)TTFT (stddev)
e40f553 — 2026-02-06 23:15:19n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)
1531781 — 2026-01-23 18:40:341.00 (n/a)1.00 (n/a)1.00 (n/a)1.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)2.70 (n/a)2.70 (n/a)2.70 (n/a)2.70 (n/a)0.00 (n/a)

llama_3.2_1b_prompt_2048_tokens_40

Commit/Date Num_Tokens (max)Num_Tokens (mean)Num_Tokens (median)Num_Tokens (min)Num_Tokens (stddev)TPS (max)TPS (mean)TPS (median)TPS (min)TPS (stddev)TTFT (max)TTFT (mean)TTFT (median)TTFT (min)TTFT (stddev)
e40f553 — 2026-02-06 23:15:19n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)
1531781 — 2026-01-23 18:40:3440.00 (n/a)40.00 (n/a)40.00 (n/a)40.00 (n/a)0.00 (n/a)4.22 (n/a)4.22 (n/a)4.22 (n/a)4.22 (n/a)0.00 (n/a)2.70 (n/a)2.70 (n/a)2.70 (n/a)2.70 (n/a)0.00 (n/a)

@github-actions
Copy link
Contributor

github-actions bot commented Feb 7, 2026

📊 Test Results for Test Example Applications

02b4f01 (2026_02_07_00_12_46)

IRONCLAD

Tested on 2026_02_07_00_12_46 at commit 02b4f01.

Test Checks TTFT (mean)TPS (mean)
llama_3.2_1b_prompt_13_tokens_1 ❌ 0/1 n/a n/a
llama_3.2_1b_prompt_13_tokens_40 ❌ 0/1 n/a n/a
llama_3.2_1b_prompt_2048_tokens_1 ❌ 0/1 n/a n/a
llama_3.2_1b_prompt_2048_tokens_40 ❌ 0/1 n/a n/a
📈 Trends (vs main branch) for Test Example Applications

02b4f01 (2026_02_07_00_12_46)

IRONCLAD Trends

llama_3.2_1b

Commit/Date Num Tokens (max)Num Tokens (mean)Num Tokens (median)Num Tokens (min)Num Tokens (stddev)TPS (max)TPS (mean)TPS (median)TPS (min)TPS (stddev)TTFT (max)TTFT (mean)TTFT (median)TTFT (min)TTFT (stddev)Total (max)Total (mean)Total (median)Total (min)Total (stddev)
130b6ea — 2025-12-05 21:33:1240.00 (+0.00%)40.00 (+0.00%)40.00 (+0.00%)40.00 (+0.00%)0.00 (n/a)4.71 (-0.42%)4.64 (-0.09%)4.64 (+0.65%)4.55 (-0.22%)0.05 (-17.66%)4.41 (-0.34%)4.39 (-0.19%)4.38 (-0.33%)4.37 (-0.15%)0.01 (-25.90%)12.96 (-0.00%)12.80 (+0.07%)12.80 (-0.23%)12.67 (+0.44%)0.09 (-21.12%)
0a6c11c — 2025-12-03 23:35:1540.00 (n/a)40.00 (n/a)40.00 (n/a)40.00 (n/a)0.00 (n/a)4.73 (n/a)4.64 (n/a)4.61 (n/a)4.56 (n/a)0.06 (n/a)4.42 (n/a)4.40 (n/a)4.40 (n/a)4.37 (n/a)0.02 (n/a)12.96 (n/a)12.79 (n/a)12.83 (n/a)12.62 (n/a)0.12 (n/a)

llama_3.2_1b_prompt_13_tokens_1

Commit/Date Num_Tokens (max)Num_Tokens (mean)Num_Tokens (median)Num_Tokens (min)Num_Tokens (stddev)TPS (max)TPS (mean)TPS (median)TPS (min)TPS (stddev)TTFT (max)TTFT (mean)TTFT (median)TTFT (min)TTFT (stddev)
02b4f01 — 2026-02-07 00:12:21n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)
1531781 — 2026-01-23 18:40:341.00 (n/a)1.00 (n/a)1.00 (n/a)1.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.61 (n/a)0.61 (n/a)0.61 (n/a)0.61 (n/a)0.00 (n/a)

llama_3.2_1b_prompt_13_tokens_40

Commit/Date Num_Tokens (max)Num_Tokens (mean)Num_Tokens (median)Num_Tokens (min)Num_Tokens (stddev)TPS (max)TPS (mean)TPS (median)TPS (min)TPS (stddev)TTFT (max)TTFT (mean)TTFT (median)TTFT (min)TTFT (stddev)
02b4f01 — 2026-02-07 00:12:21n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)
1531781 — 2026-01-23 18:40:3440.00 (n/a)40.00 (n/a)40.00 (n/a)40.00 (n/a)0.00 (n/a)4.34 (n/a)4.34 (n/a)4.34 (n/a)4.34 (n/a)0.00 (n/a)0.61 (n/a)0.61 (n/a)0.61 (n/a)0.61 (n/a)0.00 (n/a)

llama_3.2_1b_prompt_2048_tokens_1

Commit/Date Num_Tokens (max)Num_Tokens (mean)Num_Tokens (median)Num_Tokens (min)Num_Tokens (stddev)TPS (max)TPS (mean)TPS (median)TPS (min)TPS (stddev)TTFT (max)TTFT (mean)TTFT (median)TTFT (min)TTFT (stddev)
02b4f01 — 2026-02-07 00:12:21n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)
1531781 — 2026-01-23 18:40:341.00 (n/a)1.00 (n/a)1.00 (n/a)1.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)2.70 (n/a)2.70 (n/a)2.70 (n/a)2.70 (n/a)0.00 (n/a)

llama_3.2_1b_prompt_2048_tokens_40

Commit/Date Num_Tokens (max)Num_Tokens (mean)Num_Tokens (median)Num_Tokens (min)Num_Tokens (stddev)TPS (max)TPS (mean)TPS (median)TPS (min)TPS (stddev)TTFT (max)TTFT (mean)TTFT (median)TTFT (min)TTFT (stddev)
02b4f01 — 2026-02-07 00:12:21n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)
1531781 — 2026-01-23 18:40:3440.00 (n/a)40.00 (n/a)40.00 (n/a)40.00 (n/a)0.00 (n/a)4.22 (n/a)4.22 (n/a)4.22 (n/a)4.22 (n/a)0.00 (n/a)2.70 (n/a)2.70 (n/a)2.70 (n/a)2.70 (n/a)0.00 (n/a)

@github-actions
Copy link
Contributor

github-actions bot commented Feb 7, 2026

📊 Test Results for Small Benchmark/Test Suite

02b4f01 (2026_02_07_00_22_48)

IRONCLAD

Tested on 2026_02_07_00_22_48 at commit 02b4f01.

Test Checks Latency (mean)Bandwidth (mean)GFLOP/s (mean)
axpy_1_cols_2_channels_2048_tile_2048_3.0_0 ✅ 5/5 164.38 0.08 n/a
axpy_2_cols_2_channels_2048_tile_1024_3.0_0 ✅ 5/5 196.06 0.06 n/a
axpy_4_cols_2_channels_2048_tile_512_3.0_0 ✅ 5/5 174.54 0.07 n/a
axpy_8_cols_2_channels_2048_tile_256_3.0_0 ✅ 5/5 195.48 0.06 n/a
dequant_1_cols_1_channels_2048_tile_2048_0 ✅ 5/5 143.58 0.04 n/a
dequant_1_cols_2_channels_2048_tile_1024_0 ✅ 5/5 129.18 0.04 n/a
dequant_2_cols_1_channels_2048_tile_1024_0 ✅ 5/5 148.56 0.04 n/a
dequant_2_cols_2_channels_2048_tile_512_0 ✅ 5/5 158.00 0.03 n/a
dequant_4_cols_1_channels_2048_tile_512_0 ✅ 5/5 152.96 0.04 n/a
dequant_4_cols_2_channels_2048_tile_256_0 ✅ 5/5 197.90 0.03 n/a
dequant_8_cols_1_channels_2048_tile_256_0 ✅ 5/5 189.30 0.03 n/a
dequant_8_cols_2_channels_2048_tile_128_0 ✅ 5/5 203.58 0.03 n/a
eltwise_add_1_cols_2_channels_2048_tile_2048 ✅ 5/5 119.40 0.10 n/a
eltwise_add_2_cols_2_channels_2048_tile_1024 ✅ 5/5 152.10 0.09 n/a
eltwise_add_4_cols_2_channels_2048_tile_512 ✅ 5/5 150.02 0.08 n/a
eltwise_add_8_cols_2_channels_2048_tile_256 ✅ 5/5 238.36 0.06 n/a
eltwise_mul_1_cols_2_channels_2048_tile_2048 ✅ 5/5 156.84 0.08 n/a
eltwise_mul_2_cols_2_channels_2048_tile_1024 ✅ 5/5 193.98 0.06 n/a
eltwise_mul_4_cols_2_channels_2048_tile_512 ✅ 5/5 184.30 0.07 n/a
eltwise_mul_8_cols_2_channels_2048_tile_256 ✅ 5/5 242.14 0.06 n/a
gelu_1_cols_1_channels_2048_tile_2048 ✅ 5/5 174.62 0.05 n/a
gelu_1_cols_2_channels_2048_tile_1024 ✅ 5/5 168.52 0.05 n/a
gelu_2_cols_1_channels_2048_tile_1024 ✅ 5/5 172.46 0.05 n/a
gelu_2_cols_2_channels_2048_tile_512 ✅ 5/5 150.24 0.05 n/a
gelu_4_cols_1_channels_2048_tile_512 ✅ 5/5 201.78 0.04 n/a
gelu_4_cols_2_channels_2048_tile_256 ✅ 5/5 176.32 0.05 n/a
gelu_8_cols_1_channels_2048_tile_256 ✅ 5/5 164.16 0.05 n/a
gelu_8_cols_2_channels_2048_tile_128 ✅ 5/5 169.82 0.05 n/a
gemm_1792x896x1152_64x32x48_8cols_ccolmaj ✅ 5/5 2329.46 4.09 n/a
gemm_192x384x64_48x96x16_4cols ✅ 5/5 237.54 0.98 n/a
gemm_192x384x64_48x96x16_4cols_bcolmaj_ccolmaj ✅ 5/5 296.44 0.82 n/a
gemm_2048x2048x2048_64x64x64_1cols ✅ 5/5 48552.30 0.52 n/a
gemm_2048x2048x2048_64x64x64_2cols_bcolmaj ✅ 5/5 27718.36 0.91 n/a
gemm_2048x2048x2048_64x64x64_8cols_bcolmaj_ccolmaj ✅ 5/5 7546.08 3.34 n/a
gemm_384x1536x1792_32x48x64_4cols_bcolmaj ✅ 5/5 2169.64 3.79 n/a
gemm_896x1792x640_32x64x80_8cols_ccolmaj ✅ 5/5 1272.38 5.35 n/a
layer_norm_1_cols_1_channels_2048_tile_2048 ✅ 5/5 153.50 0.06 n/a
layer_norm_1_cols_2_channels_2048_tile_1024 ✅ 5/5 175.36 0.05 n/a
layer_norm_2_cols_1_channels_2048_tile_1024 ✅ 5/5 143.52 0.06 n/a
layer_norm_2_cols_2_channels_2048_tile_512 ✅ 5/5 189.94 0.05 n/a
layer_norm_4_cols_1_channels_2048_tile_512 ✅ 5/5 206.96 0.04 n/a
layer_norm_4_cols_2_channels_2048_tile_256 ✅ 5/5 208.18 0.04 n/a
layer_norm_8_cols_1_channels_2048_tile_256 ✅ 5/5 176.28 0.05 n/a
layer_norm_8_cols_2_channels_2048_tile_128 ✅ 5/5 251.20 0.04 n/a
matrix_vector_mul_128x128_32tsi_128tso_1col0 ✅ 5/5 n/a 0.24 n/a
matrix_vector_mul_2048x8192_1tsi_1024tso_2col0 ✅ 5/5 n/a 24.16 n/a
matrix_vector_mul_2048x8192_1tsi_2048tso_1col0 ✅ 5/5 n/a 11.94 n/a
matrix_vector_mul_2048x8192_1tsi_256tso_8col0 ✅ 5/5 n/a 43.37 n/a
matrix_vector_mul_2048x8192_1tsi_512tso_4col0 ✅ 5/5 n/a 39.88 n/a
matrix_vector_mul_8192x2048_4tsi_1024tso_1col0 ✅ 5/5 n/a 13.21 n/a
matrix_vector_mul_8192x2048_4tsi_1024tso_2col0 ✅ 5/5 n/a 24.01 n/a
matrix_vector_mul_8192x2048_4tsi_1024tso_4col0 ✅ 5/5 n/a 39.60 n/a
matrix_vector_mul_8192x2048_4tsi_1024tso_8col0 ✅ 5/5 n/a 42.88 n/a
mem_copy_16_cores_2_chans_2048_tile_128_False0 ✅ 5/5 218.28 0.04 n/a
mem_copy_1_cores_1_chans_2048_tile_2048_False0 ✅ 5/5 159.08 0.05 n/a
mem_copy_2_cores_1_chans_2048_tile_1024_False0 ✅ 5/5 164.04 0.05 n/a
mem_copy_2_cores_2_chans_2048_tile_1024_False0 ✅ 5/5 177.16 0.05 n/a
mem_copy_4_cores_1_chans_2048_tile_512_False0 ✅ 5/5 168.94 0.05 n/a
mem_copy_4_cores_2_chans_2048_tile_512_False0 ✅ 5/5 159.00 0.05 n/a
mem_copy_8_cores_1_chans_2048_tile_256_False0 ✅ 5/5 175.18 0.05 n/a
mem_copy_8_cores_2_chans_2048_tile_256_False0 ✅ 5/5 163.84 0.05 n/a
mha0 ✅ 5/5 47597.74 0.18 n/a
relu_1_cols_1_channels_2048_tile_2048 ✅ 5/5 146.66 0.06 n/a
relu_2_cols_1_channels_2048_tile_1024 ✅ 5/5 149.52 0.06 n/a
relu_4_cols_1_channels_2048_tile_512 ✅ 5/5 177.80 0.05 n/a
relu_8_cols_1_channels_2048_tile_256 ✅ 5/5 213.32 0.04 n/a
rms_norm_1_cols_1_channels_2048_tile_2048 ✅ 5/5 153.32 0.05 n/a
rms_norm_1_cols_2_channels_2048_tile_1024 ✅ 5/5 140.94 0.06 n/a
rms_norm_2_cols_1_channels_2048_tile_1024 ✅ 5/5 146.86 0.06 n/a
rms_norm_2_cols_2_channels_2048_tile_512 ✅ 5/5 198.16 0.04 n/a
rms_norm_4_cols_1_channels_2048_tile_512 ✅ 5/5 154.64 0.05 n/a
rms_norm_4_cols_2_channels_2048_tile_256 ✅ 5/5 167.36 0.05 n/a
rms_norm_8_cols_1_channels_2048_tile_256 ✅ 5/5 210.20 0.04 n/a
rms_norm_8_cols_2_channels_2048_tile_128 ✅ 5/5 241.46 0.04 n/a
rope_1c_32rows_512cols_32arows_0m ✅ 5/5 165.50 0.62 n/a
rope_1c_32rows_512cols_8arows_0m ✅ 5/5 136.46 0.55 n/a
rope_2c_32rows_512cols_32arows_0m ✅ 5/5 173.60 0.59 n/a
rope_2c_32rows_512cols_8arows_0m ✅ 5/5 204.04 0.44 n/a
rope_8c_32rows_512cols_32arows_0m ✅ 5/5 190.06 0.55 n/a
rope_8c_32rows_512cols_8arows_0m ✅ 5/5 202.92 0.36 n/a
sigmoid_1_cols_1_channels_2048_tile_2048 ✅ 5/5 127.22 0.07 n/a
sigmoid_2_cols_1_channels_2048_tile_1024 ✅ 5/5 136.32 0.06 n/a
sigmoid_4_cols_1_channels_2048_tile_512 ✅ 5/5 117.16 0.07 n/a
sigmoid_8_cols_1_channels_2048_tile_256 ✅ 5/5 139.14 0.06 n/a
silu_1_cols_1_channels_2048_tile_2048 ✅ 5/5 150.38 0.06 n/a
silu_2_cols_1_channels_2048_tile_1024 ✅ 5/5 165.56 0.05 n/a
silu_4_cols_1_channels_2048_tile_512 ✅ 5/5 140.20 0.06 n/a
silu_8_cols_1_channels_2048_tile_256 ✅ 5/5 146.06 0.06 n/a
softmax_2_cols_2_channels_32768_tile_1024 ✅ 5/5 146.10 0.92 n/a
softmax_2_cols_2_channels_32768_tile_512 ✅ 5/5 164.32 0.84 n/a
softmax_4_cols_4_channels_32768_tile_2048 ✅ 5/5 192.58 0.75 n/a
swiglu_decode_1x2048x2048_0 ✅ 5/5 n/a n/a n/a
swiglu_prefill_256x2048x2048_0 ❌ 0/5 n/a n/a n/a
tanh_1_cols_1_channels_2048_tile_2048 ✅ 5/5 129.90 0.07 n/a
tanh_2_cols_1_channels_2048_tile_1024 ✅ 5/5 133.44 0.07 n/a
tanh_4_cols_1_channels_2048_tile_512 ✅ 5/5 144.36 0.06 n/a
tanh_8_cols_1_channels_2048_tile_256 ✅ 5/5 181.62 0.05 n/a
transpose_2048_M_64_N_1_cols_1_channels_64_m_64_n_8_s0 ✅ 5/5 180.64 2.93 n/a
transpose_2048_M_64_N_1_cols_2_channels_64_m_64_n_8_s0 ✅ 5/5 152.62 3.53 n/a
weighted_rms_norm_1_cols_2_channels_2048_weights_2048 ✅ 5/5 167.94 0.08 n/a
weighted_rms_norm_2_cols_2_channels_2048_weights_1024 ✅ 5/5 170.10 0.06 n/a
weighted_rms_norm_4_cols_2_channels_2048_weights_512 ✅ 5/5 207.22 0.05 n/a
weighted_rms_norm_8_cols_2_channels_2048_weights_256 ✅ 5/5 193.78 0.05 n/a
📈 Trends (vs main branch) for Small Benchmark/Test Suite

02b4f01 (2026_02_07_00_22_48)

IRONCLAD Trends

axpy_1_cols_2_channels_2048_tile_2048_3.0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.13 (-1.96%)0.11 (+2.52%)0.10 (+1.86%)0.08 (-1.91%)0.02 (-1.72%)162.40 (+1.95%)120.96 (-2.36%)117.00 (-1.85%)94.40 (+2.05%)23.11 (+4.09%)
0a6c11c — 2025-12-04 00:39:100.13 (n/a)0.10 (n/a)0.10 (n/a)0.08 (n/a)0.02 (n/a)159.30 (n/a)123.88 (n/a)119.20 (n/a)92.50 (n/a)22.21 (n/a)

axpy_1_cols_2_channels_2048_tile_2048_3.0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.10 (-33.37%)0.08 (-39.39%)0.07 (-34.85%)0.06 (-38.64%)0.02 (-41.80%)193.60 (+62.96%)164.38 (+63.89%)169.50 (+53.53%)117.10 (+50.13%)28.51 (+39.15%)
1531781 — 2026-01-23 18:49:240.16 (n/a)0.13 (n/a)0.11 (n/a)0.10 (n/a)0.03 (n/a)118.80 (n/a)100.30 (n/a)110.40 (n/a)78.00 (n/a)20.49 (n/a)

axpy_2_cols_2_channels_2048_tile_1024_3.0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.11 (-26.77%)0.09 (-9.09%)0.09 (-10.81%)0.08 (+8.25%)0.01 (-55.42%)152.00 (-7.60%)135.46 (+6.28%)142.80 (+12.18%)115.00 (+36.58%)15.03 (-41.36%)
0a6c11c — 2025-12-04 00:39:100.15 (n/a)0.10 (n/a)0.10 (n/a)0.07 (n/a)0.02 (n/a)164.50 (n/a)127.46 (n/a)127.30 (n/a)84.20 (n/a)25.64 (n/a)

axpy_2_cols_2_channels_2048_tile_1024_3.0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.09 (-37.27%)0.06 (-47.35%)0.06 (-53.43%)0.05 (-38.38%)0.01 (-44.55%)254.60 (+62.27%)196.06 (+87.69%)193.80 (+114.62%)140.20 (+59.50%)40.64 (+37.67%)
1531781 — 2026-01-23 18:49:240.14 (n/a)0.12 (n/a)0.14 (n/a)0.08 (n/a)0.03 (n/a)156.90 (n/a)104.46 (n/a)90.30 (n/a)87.90 (n/a)29.52 (n/a)

axpy_4_cols_2_channels_2048_tile_512_3.0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.12 (-10.21%)0.10 (-8.41%)0.10 (-6.20%)0.07 (+3.52%)0.02 (-23.48%)166.70 (-3.36%)128.78 (+7.33%)121.80 (+6.65%)103.90 (+11.36%)22.95 (-18.82%)
0a6c11c — 2025-12-04 00:39:100.13 (n/a)0.11 (n/a)0.11 (n/a)0.07 (n/a)0.02 (n/a)172.50 (n/a)119.98 (n/a)114.20 (n/a)93.30 (n/a)28.27 (n/a)

axpy_4_cols_2_channels_2048_tile_512_3.0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.09 (-41.84%)0.07 (-47.31%)0.07 (-52.21%)0.05 (-40.81%)0.01 (-43.93%)226.50 (+68.90%)174.54 (+88.81%)174.50 (+109.23%)135.20 (+72.01%)36.40 (+55.66%)
1531781 — 2026-01-23 18:49:240.16 (n/a)0.14 (n/a)0.15 (n/a)0.09 (n/a)0.03 (n/a)134.10 (n/a)92.44 (n/a)83.40 (n/a)78.60 (n/a)23.38 (n/a)

axpy_8_cols_2_channels_2048_tile_256_3.0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.10 (-16.19%)0.08 (-10.40%)0.09 (-2.09%)0.07 (-17.38%)0.01 (-10.61%)184.40 (+21.00%)153.18 (+11.99%)144.10 (+2.20%)127.10 (+19.34%)21.26 (+34.76%)
0a6c11c — 2025-12-04 00:39:100.12 (n/a)0.09 (n/a)0.09 (n/a)0.08 (n/a)0.01 (n/a)152.40 (n/a)136.78 (n/a)141.00 (n/a)106.50 (n/a)15.78 (n/a)

axpy_8_cols_2_channels_2048_tile_256_3.0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.08 (-53.81%)0.06 (-55.26%)0.06 (-62.02%)0.06 (-31.26%)0.01 (-74.29%)213.60 (+45.50%)195.48 (+111.93%)210.50 (+163.12%)156.60 (+116.60%)24.51 (-21.12%)
1531781 — 2026-01-23 18:49:240.17 (n/a)0.14 (n/a)0.15 (n/a)0.08 (n/a)0.03 (n/a)146.80 (n/a)92.24 (n/a)80.00 (n/a)72.30 (n/a)31.07 (n/a)

dequant_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.05 (-18.83%)0.04 (-5.21%)0.04 (-2.69%)0.03 (+37.10%)0.00 (-53.98%)160.40 (-27.06%)141.12 (-0.91%)136.40 (+2.79%)114.30 (+23.17%)17.37 (-59.06%)
0a6c11c — 2025-12-04 00:39:100.06 (n/a)0.04 (n/a)0.04 (n/a)0.02 (n/a)0.01 (n/a)219.90 (n/a)142.42 (n/a)132.70 (n/a)92.80 (n/a)42.43 (n/a)

dequant_1_cols_1_channels_2048_tile_2048_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.04 (-27.66%)0.04 (-20.21%)0.04 (-15.42%)0.03 (-26.39%)0.00 (-37.47%)170.50 (+35.86%)143.58 (+24.77%)143.80 (+18.26%)121.80 (+38.25%)18.45 (+19.80%)
1531781 — 2026-01-23 18:49:240.06 (n/a)0.05 (n/a)0.04 (n/a)0.04 (n/a)0.01 (n/a)125.50 (n/a)115.08 (n/a)121.60 (n/a)88.10 (n/a)15.40 (n/a)

dequant_1_cols_2_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.06 (-13.28%)0.05 (-8.93%)0.06 (-13.59%)0.04 (+4.63%)0.01 (-39.10%)133.60 (-4.43%)106.08 (+5.85%)95.10 (+15.69%)87.90 (+15.35%)17.90 (-31.97%)
0a6c11c — 2025-12-04 00:39:100.07 (n/a)0.06 (n/a)0.06 (n/a)0.04 (n/a)0.01 (n/a)139.80 (n/a)100.22 (n/a)82.20 (n/a)76.20 (n/a)26.31 (n/a)

dequant_1_cols_2_channels_2048_tile_1024_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.05 (-21.73%)0.04 (-14.57%)0.04 (-17.36%)0.04 (+11.49%)0.01 (-45.33%)146.90 (-10.26%)129.18 (+12.66%)134.70 (+21.02%)97.80 (+27.84%)19.38 (-38.74%)
1531781 — 2026-01-23 18:49:240.07 (n/a)0.05 (n/a)0.05 (n/a)0.03 (n/a)0.01 (n/a)163.70 (n/a)114.66 (n/a)111.30 (n/a)76.50 (n/a)31.63 (n/a)

dequant_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.07 (+5.66%)0.05 (+1.21%)0.05 (+0.61%)0.04 (-15.02%)0.01 (+58.42%)143.40 (+17.64%)101.02 (+1.49%)97.60 (-0.61%)78.90 (-5.40%)22.56 (+78.52%)
0a6c11c — 2025-12-04 00:39:100.06 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)121.90 (n/a)99.54 (n/a)98.20 (n/a)83.40 (n/a)12.64 (n/a)

dequant_2_cols_1_channels_2048_tile_1024_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.05 (-3.74%)0.04 (-21.84%)0.04 (-28.66%)0.03 (-21.25%)0.01 (+33.81%)187.60 (+26.93%)148.56 (+31.31%)149.30 (+40.19%)105.40 (+3.94%)34.35 (+75.12%)
1531781 — 2026-01-23 18:49:240.05 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)147.80 (n/a)113.14 (n/a)106.50 (n/a)101.40 (n/a)19.61 (n/a)

dequant_2_cols_2_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.05 (+17.06%)0.04 (+8.98%)0.05 (+21.37%)0.03 (-11.65%)0.01 (+301.89%)159.70 (+13.18%)126.32 (-5.68%)109.00 (-17.67%)106.80 (-14.56%)22.48 (+280.15%)
0a6c11c — 2025-12-04 00:39:100.04 (n/a)0.04 (n/a)0.04 (n/a)0.04 (n/a)0.00 (n/a)141.10 (n/a)133.92 (n/a)132.40 (n/a)125.00 (n/a)5.91 (n/a)

dequant_2_cols_2_channels_2048_tile_512_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.04 (-20.48%)0.03 (-23.88%)0.03 (-34.02%)0.03 (-20.36%)0.01 (-31.40%)198.80 (+25.58%)158.00 (+30.17%)160.90 (+51.51%)123.50 (+25.76%)28.21 (+7.99%)
1531781 — 2026-01-23 18:49:240.05 (n/a)0.04 (n/a)0.05 (n/a)0.03 (n/a)0.01 (n/a)158.30 (n/a)121.38 (n/a)106.20 (n/a)98.20 (n/a)26.13 (n/a)

dequant_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.06 (+20.24%)0.05 (-3.19%)0.05 (-6.73%)0.03 (-20.69%)0.01 (+194.26%)156.10 (+26.09%)120.86 (+7.15%)114.90 (+7.18%)88.30 (-16.85%)24.40 (+214.82%)
0a6c11c — 2025-12-04 00:39:100.05 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.00 (n/a)123.80 (n/a)112.80 (n/a)107.20 (n/a)106.20 (n/a)7.75 (n/a)

dequant_4_cols_1_channels_2048_tile_512_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.04 (-28.70%)0.04 (-27.42%)0.03 (-32.76%)0.03 (-18.31%)0.01 (-41.19%)201.70 (+22.39%)152.96 (+34.62%)153.50 (+48.74%)119.00 (+40.33%)32.58 (+0.07%)
1531781 — 2026-01-23 18:49:240.06 (n/a)0.05 (n/a)0.05 (n/a)0.03 (n/a)0.01 (n/a)164.80 (n/a)113.62 (n/a)103.20 (n/a)84.80 (n/a)32.55 (n/a)

dequant_4_cols_2_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.05 (-6.09%)0.04 (-8.04%)0.04 (-12.46%)0.03 (-0.41%)0.01 (-20.19%)168.10 (+0.42%)135.74 (+7.44%)132.20 (+14.26%)107.40 (+6.44%)22.91 (-12.67%)
0a6c11c — 2025-12-04 00:39:100.05 (n/a)0.04 (n/a)0.05 (n/a)0.03 (n/a)0.01 (n/a)167.40 (n/a)126.34 (n/a)115.70 (n/a)100.90 (n/a)26.23 (n/a)

dequant_4_cols_2_channels_2048_tile_256_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.03 (-49.99%)0.03 (-46.63%)0.02 (-56.31%)0.02 (-17.17%)0.00 (-66.60%)227.50 (+20.75%)197.90 (+75.57%)214.80 (+129.00%)159.30 (+99.87%)32.92 (-24.45%)
1531781 — 2026-01-23 18:49:240.07 (n/a)0.05 (n/a)0.06 (n/a)0.03 (n/a)0.01 (n/a)188.40 (n/a)112.72 (n/a)93.80 (n/a)79.70 (n/a)43.58 (n/a)

dequant_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.04 (-13.65%)0.04 (-13.67%)0.04 (-18.56%)0.03 (-13.52%)0.00 (-23.88%)170.40 (+15.68%)140.12 (+15.33%)135.80 (+22.78%)119.60 (+15.78%)17.94 (+1.87%)
0a6c11c — 2025-12-04 00:39:100.05 (n/a)0.04 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)147.30 (n/a)121.50 (n/a)110.60 (n/a)103.30 (n/a)17.61 (n/a)

dequant_8_cols_1_channels_2048_tile_256_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.03 (-32.16%)0.03 (-27.26%)0.03 (-28.40%)0.02 (-21.41%)0.00 (-50.14%)212.40 (+27.26%)189.30 (+36.52%)189.80 (+39.66%)173.30 (+47.36%)16.35 (-8.87%)
1531781 — 2026-01-23 18:49:240.04 (n/a)0.04 (n/a)0.04 (n/a)0.03 (n/a)0.00 (n/a)166.90 (n/a)138.66 (n/a)135.90 (n/a)117.60 (n/a)17.94 (n/a)

dequant_8_cols_2_channels_2048_tile_128

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.05 (+17.21%)0.03 (-1.70%)0.03 (-3.04%)0.03 (-16.20%)0.01 (+95.59%)198.20 (+19.33%)160.00 (+4.92%)166.20 (+3.17%)106.50 (-14.66%)29.86 (+87.19%)
0a6c11c — 2025-12-04 00:39:100.04 (n/a)0.03 (n/a)0.03 (n/a)0.03 (n/a)0.00 (n/a)166.10 (n/a)152.50 (n/a)161.10 (n/a)124.80 (n/a)15.95 (n/a)

dequant_8_cols_2_channels_2048_tile_128_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.04 (-26.64%)0.03 (-31.19%)0.02 (-40.80%)0.02 (-3.68%)0.01 (-42.49%)238.70 (+3.83%)203.58 (+39.82%)221.30 (+68.93%)144.70 (+36.25%)37.36 (-23.80%)
1531781 — 2026-01-23 18:49:240.05 (n/a)0.04 (n/a)0.04 (n/a)0.02 (n/a)0.01 (n/a)229.90 (n/a)145.60 (n/a)131.00 (n/a)106.20 (n/a)49.03 (n/a)

eltwise_add_1_cols_2_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.12 (-28.35%)0.10 (-14.32%)0.11 (+4.19%)0.09 (+8.93%)0.01 (-68.68%)143.10 (-8.15%)119.40 (+8.76%)115.60 (-4.07%)103.00 (+39.57%)14.88 (-56.72%)
1531781 — 2026-01-23 18:49:240.17 (n/a)0.12 (n/a)0.10 (n/a)0.08 (n/a)0.04 (n/a)155.80 (n/a)109.78 (n/a)120.50 (n/a)73.80 (n/a)34.38 (n/a)

eltwise_add_2_cols_2_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.11 (-26.45%)0.09 (-31.11%)0.09 (-33.59%)0.05 (-39.97%)0.02 (-7.97%)244.40 (+66.60%)152.10 (+50.30%)138.00 (+50.49%)110.60 (+35.87%)53.64 (+106.34%)
1531781 — 2026-01-23 18:49:240.15 (n/a)0.13 (n/a)0.13 (n/a)0.08 (n/a)0.03 (n/a)146.70 (n/a)101.20 (n/a)91.70 (n/a)81.40 (n/a)26.00 (n/a)

eltwise_add_4_cols_2_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.11 (-34.41%)0.08 (-42.00%)0.08 (-45.09%)0.06 (-40.79%)0.02 (-29.04%)196.00 (+68.97%)150.02 (+73.51%)147.30 (+82.08%)110.90 (+52.54%)30.55 (+77.39%)
1531781 — 2026-01-23 18:49:240.17 (n/a)0.15 (n/a)0.15 (n/a)0.11 (n/a)0.02 (n/a)116.00 (n/a)86.46 (n/a)80.90 (n/a)72.70 (n/a)17.22 (n/a)

eltwise_add_8_cols_2_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.08 (-51.26%)0.06 (-58.09%)0.06 (-57.08%)0.03 (-74.52%)0.02 (-5.40%)429.60 (+292.33%)238.36 (+164.84%)200.30 (+132.91%)154.50 (+105.18%)109.66 (+716.15%)
1531781 — 2026-01-23 18:49:240.16 (n/a)0.14 (n/a)0.14 (n/a)0.11 (n/a)0.02 (n/a)109.50 (n/a)90.00 (n/a)86.00 (n/a)75.30 (n/a)13.44 (n/a)

eltwise_mul_1_cols_2_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.09 (-35.06%)0.08 (-37.31%)0.08 (-37.26%)0.06 (-39.74%)0.01 (-24.75%)196.00 (+65.96%)156.84 (+60.24%)147.60 (+59.40%)139.20 (+53.98%)22.55 (+95.27%)
1531781 — 2026-01-23 18:49:240.14 (n/a)0.13 (n/a)0.13 (n/a)0.10 (n/a)0.01 (n/a)118.10 (n/a)97.88 (n/a)92.60 (n/a)90.40 (n/a)11.55 (n/a)

eltwise_mul_2_cols_2_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.07 (-46.04%)0.06 (-47.62%)0.07 (-50.38%)0.05 (-42.28%)0.01 (-52.63%)225.30 (+73.31%)193.98 (+89.99%)188.10 (+101.61%)168.80 (+85.29%)25.14 (+52.07%)
1531781 — 2026-01-23 18:49:240.13 (n/a)0.12 (n/a)0.13 (n/a)0.09 (n/a)0.02 (n/a)130.00 (n/a)102.10 (n/a)93.30 (n/a)91.10 (n/a)16.53 (n/a)

eltwise_mul_4_cols_2_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.08 (-51.45%)0.07 (-53.15%)0.07 (-53.14%)0.05 (-49.17%)0.01 (-47.05%)224.40 (+96.84%)184.30 (+113.85%)173.80 (+113.25%)152.80 (+105.93%)32.86 (+107.32%)
1531781 — 2026-01-23 18:49:240.17 (n/a)0.15 (n/a)0.15 (n/a)0.11 (n/a)0.02 (n/a)114.00 (n/a)86.18 (n/a)81.50 (n/a)74.20 (n/a)15.85 (n/a)

eltwise_mul_8_cols_2_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.06 (-49.86%)0.06 (-50.36%)0.06 (-47.79%)0.03 (-63.76%)0.02 (-20.91%)427.20 (+175.97%)242.14 (+115.66%)200.30 (+91.49%)190.50 (+99.48%)103.57 (+329.13%)
1531781 — 2026-01-23 18:49:240.13 (n/a)0.11 (n/a)0.12 (n/a)0.08 (n/a)0.02 (n/a)154.80 (n/a)112.28 (n/a)104.60 (n/a)95.50 (n/a)24.14 (n/a)

gelu_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.07 (-34.86%)0.05 (-50.04%)0.05 (-56.65%)0.03 (-58.97%)0.01 (-15.34%)247.70 (+143.80%)174.62 (+108.38%)167.10 (+130.80%)110.90 (+53.60%)49.40 (+213.62%)
1531781 — 2026-01-23 18:49:240.11 (n/a)0.10 (n/a)0.11 (n/a)0.08 (n/a)0.02 (n/a)101.60 (n/a)83.80 (n/a)72.40 (n/a)72.20 (n/a)15.75 (n/a)

gelu_1_cols_2_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.06 (-52.94%)0.05 (-53.25%)0.05 (-56.31%)0.04 (-47.05%)0.01 (-65.31%)207.70 (+88.82%)168.52 (+109.97%)158.10 (+128.80%)141.10 (+112.50%)26.29 (+41.81%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.11 (n/a)0.12 (n/a)0.07 (n/a)0.02 (n/a)110.00 (n/a)80.26 (n/a)69.10 (n/a)66.40 (n/a)18.54 (n/a)

gelu_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.06 (-49.25%)0.05 (-53.23%)0.05 (-61.49%)0.04 (-30.88%)0.01 (-68.29%)199.30 (+44.63%)172.46 (+102.85%)177.10 (+159.68%)133.50 (+96.90%)27.71 (-8.36%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.10 (n/a)0.12 (n/a)0.06 (n/a)0.03 (n/a)137.80 (n/a)85.02 (n/a)68.20 (n/a)67.80 (n/a)30.24 (n/a)

gelu_2_cols_2_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.06 (-51.33%)0.05 (-50.32%)0.06 (-53.18%)0.05 (-34.82%)0.00 (-81.39%)165.50 (+53.53%)150.24 (+95.63%)148.50 (+113.67%)137.80 (+105.37%)10.06 (-42.18%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.11 (n/a)0.12 (n/a)0.08 (n/a)0.02 (n/a)107.80 (n/a)76.80 (n/a)69.50 (n/a)67.10 (n/a)17.40 (n/a)

gelu_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.06 (-57.15%)0.04 (-59.65%)0.04 (-62.45%)0.03 (-47.07%)0.01 (-64.73%)289.20 (+88.90%)201.78 (+135.83%)184.50 (+166.23%)147.10 (+133.49%)56.20 (+48.15%)
1531781 — 2026-01-23 18:49:240.13 (n/a)0.11 (n/a)0.12 (n/a)0.05 (n/a)0.03 (n/a)153.10 (n/a)85.56 (n/a)69.30 (n/a)63.00 (n/a)37.94 (n/a)

gelu_4_cols_2_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.06 (-44.47%)0.05 (-53.03%)0.05 (-54.40%)0.03 (-53.76%)0.01 (-24.63%)238.90 (+116.20%)176.32 (+120.95%)154.50 (+119.15%)126.10 (+80.14%)52.44 (+199.94%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.11 (n/a)0.12 (n/a)0.07 (n/a)0.02 (n/a)110.50 (n/a)79.80 (n/a)70.50 (n/a)70.00 (n/a)17.48 (n/a)

gelu_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.05 (-53.17%)0.05 (-49.37%)0.05 (-53.70%)0.04 (+4.11%)0.01 (-83.23%)200.00 (-3.94%)164.16 (+66.25%)155.20 (+115.86%)150.00 (+113.68%)20.87 (-65.90%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.10 (n/a)0.11 (n/a)0.04 (n/a)0.03 (n/a)208.20 (n/a)98.74 (n/a)71.90 (n/a)70.20 (n/a)61.20 (n/a)

gelu_8_cols_2_channels_2048_tile_128

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.06 (-38.67%)0.05 (-45.98%)0.05 (-54.03%)0.04 (-31.34%)0.01 (-54.51%)194.30 (+45.65%)169.82 (+81.74%)179.10 (+117.62%)133.50 (+63.00%)23.74 (+5.66%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.09 (n/a)0.10 (n/a)0.06 (n/a)0.02 (n/a)133.40 (n/a)93.44 (n/a)82.30 (n/a)81.90 (n/a)22.47 (n/a)

gemm_1792x896x1152_64x32x48_8cols_ccolmaj

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
02b4f01 — 2026-02-07 00:16:314.86 (-5.52%)4.09 (-16.49%)4.07 (-18.32%)3.42 (-19.96%)0.52 (+44.99%)2750.90 (+24.94%)2329.46 (+20.70%)2307.90 (+22.44%)1933.30 (+5.84%)290.77 (+88.75%)1913.47 (-5.52%)1608.17 (-16.49%)1602.92 (-18.32%)1344.80 (-19.96%)202.94 (+44.99%)
1531781 — 2026-01-23 18:49:245.15 (n/a)4.90 (n/a)4.99 (n/a)4.27 (n/a)0.36 (n/a)2201.70 (n/a)1929.94 (n/a)1885.00 (n/a)1826.70 (n/a)154.05 (n/a)2025.19 (n/a)1925.76 (n/a)1962.51 (n/a)1680.20 (n/a)139.97 (n/a)

gemm_192x384x64_48x96x16_4cols

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
02b4f01 — 2026-02-07 00:16:311.23 (-47.85%)0.98 (-47.28%)1.00 (-43.55%)0.65 (-50.89%)0.22 (-45.13%)341.70 (+103.64%)237.54 (+91.19%)220.30 (+77.09%)179.80 (+91.89%)62.14 (+121.05%)52.50 (-47.85%)41.62 (-47.28%)42.84 (-43.55%)27.62 (-50.89%)9.18 (-45.13%)
1531781 — 2026-01-23 18:49:242.36 (n/a)1.85 (n/a)1.78 (n/a)1.32 (n/a)0.39 (n/a)167.80 (n/a)124.24 (n/a)124.40 (n/a)93.70 (n/a)28.11 (n/a)100.67 (n/a)78.94 (n/a)75.88 (n/a)56.25 (n/a)16.73 (n/a)

gemm_192x384x64_48x96x16_4cols_bcolmaj_ccolmaj

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
02b4f01 — 2026-02-07 00:16:311.06 (-52.26%)0.82 (-59.92%)0.96 (-56.37%)0.52 (-67.26%)0.27 (+1.70%)422.00 (+205.35%)296.44 (+171.47%)230.30 (+129.15%)209.10 (+109.52%)108.50 (+556.07%)45.13 (-52.26%)35.18 (-59.92%)40.98 (-56.37%)22.36 (-67.26%)11.45 (+1.70%)
1531781 — 2026-01-23 18:49:242.22 (n/a)2.06 (n/a)2.20 (n/a)1.60 (n/a)0.26 (n/a)138.20 (n/a)109.20 (n/a)100.50 (n/a)99.80 (n/a)16.54 (n/a)94.54 (n/a)87.77 (n/a)93.91 (n/a)68.30 (n/a)11.26 (n/a)

gemm_2048x2048x2048_64x64x32_8_cols_0_bcolmaj_0_ccolmaj_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)GFLOP/s (max)GFLOP/s (mean)GFLOP/s (median)GFLOP/s (min)GFLOP/s (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3a053d3 — 2025-11-26 01:12:163.72 (-0.11%)3.55 (-2.75%)3.56 (-2.75%)3.36 (-5.33%)0.12 (+76.10%)2725.23 (-0.11%)2602.52 (-2.75%)2612.11 (-2.75%)2459.89 (-5.33%)86.27 (+76.10%)6984.00 (+5.63%)6608.60 (+2.91%)6577.00 (+2.83%)6304.00 (+0.11%)221.87 (+87.15%)
9e98049 — 2025-11-25 20:50:553.72 (n/a)3.65 (n/a)3.66 (n/a)3.54 (n/a)0.07 (n/a)2728.26 (n/a)2676.06 (n/a)2686.03 (n/a)2598.29 (n/a)48.99 (n/a)6612.00 (n/a)6422.00 (n/a)6396.00 (n/a)6297.00 (n/a)118.55 (n/a)

gemm_2048x2048x2048_64x64x32_8_cols_0_bcolmaj_1_ccolmaj_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)GFLOP/s (max)GFLOP/s (mean)GFLOP/s (median)GFLOP/s (min)GFLOP/s (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3a053d3 — 2025-11-26 01:12:163.40 (-3.14%)3.38 (-0.82%)3.38 (-0.82%)3.35 (+4.93%)0.02 (-84.51%)2493.81 (-3.14%)2475.27 (-0.82%)2479.06 (-0.82%)2453.92 (+4.93%)13.33 (-84.51%)7001.00 (-4.70%)6940.80 (+0.71%)6930.00 (+0.83%)6889.00 (+3.24%)37.43 (-84.76%)
9e98049 — 2025-11-25 20:50:553.51 (n/a)3.40 (n/a)3.41 (n/a)3.19 (n/a)0.12 (n/a)2574.53 (n/a)2495.72 (n/a)2499.62 (n/a)2338.67 (n/a)86.02 (n/a)7346.00 (n/a)6892.20 (n/a)6873.00 (n/a)6673.00 (n/a)245.57 (n/a)

gemm_2048x2048x2048_64x64x32_8_cols_1_bcolmaj_0_ccolmaj_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)GFLOP/s (max)GFLOP/s (mean)GFLOP/s (median)GFLOP/s (min)GFLOP/s (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3a053d3 — 2025-11-26 01:12:163.02 (+0.52%)2.87 (-2.05%)2.86 (-2.24%)2.70 (-6.16%)0.11 (+131.66%)2216.19 (+0.52%)2104.44 (-2.05%)2094.08 (-2.24%)1980.84 (-6.16%)80.25 (+131.67%)8673.00 (+6.56%)8175.60 (+2.21%)8204.00 (+2.29%)7752.00 (-0.51%)314.03 (+145.36%)
9e98049 — 2025-11-25 20:50:553.01 (n/a)2.93 (n/a)2.92 (n/a)2.88 (n/a)0.05 (n/a)2204.81 (n/a)2148.41 (n/a)2142.13 (n/a)2110.81 (n/a)34.64 (n/a)8139.00 (n/a)7998.60 (n/a)8020.00 (n/a)7792.00 (n/a)127.99 (n/a)

gemm_2048x2048x2048_64x64x64_1cols

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
02b4f01 — 2026-02-07 00:16:310.52 (-0.33%)0.52 (-0.37%)0.52 (-0.39%)0.52 (-0.40%)0.00 (+28.93%)48610.20 (+0.40%)48552.30 (+0.38%)48551.60 (+0.39%)48506.50 (+0.34%)38.15 (+29.80%)354.18 (-0.33%)353.84 (-0.37%)353.85 (-0.39%)353.42 (-0.40%)0.28 (+28.92%)
1531781 — 2026-01-23 18:49:240.52 (n/a)0.52 (n/a)0.52 (n/a)0.52 (n/a)0.00 (n/a)48415.10 (n/a)48370.52 (n/a)48362.20 (n/a)48344.40 (n/a)29.40 (n/a)355.36 (n/a)355.17 (n/a)355.23 (n/a)354.85 (n/a)0.22 (n/a)

gemm_2048x2048x2048_64x64x64_2_cols_0_bcolmaj_0_ccolmaj_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:491.04 (+0.94%)1.03 (+0.23%)1.03 (+0.09%)1.02 (+0.07%)0.00 (+197.51%)24576.00 (-0.07%)24485.82 (-0.23%)24530.40 (-0.09%)24266.40 (-0.94%)111.26 (+194.35%)
0a6c11c — 2025-12-04 00:39:101.03 (n/a)1.03 (n/a)1.03 (n/a)1.02 (n/a)0.00 (n/a)24594.30 (n/a)24542.14 (n/a)24551.90 (n/a)24495.50 (n/a)37.80 (n/a)

gemm_2048x2048x2048_64x64x64_2_cols_0_bcolmaj_0_ccolmaj_0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
a4b6ffe — 2026-01-05 19:24:051.04 (+0.28%)1.03 (-0.15%)1.03 (+0.05%)1.02 (-0.52%)0.01 (+42.72%)24719.80 (+0.52%)24512.42 (+0.16%)24529.20 (-0.05%)24092.60 (-0.28%)252.92 (+43.15%)713.08 (+0.28%)700.92 (-0.15%)700.38 (+0.05%)694.98 (-0.52%)7.30 (+42.71%)
50ad37c — 2025-12-29 20:35:531.04 (n/a)1.03 (n/a)1.03 (n/a)1.02 (n/a)0.01 (n/a)24592.30 (n/a)24473.70 (n/a)24540.90 (n/a)24160.80 (n/a)176.68 (n/a)711.06 (n/a)702.00 (n/a)700.05 (n/a)698.59 (n/a)5.12 (n/a)

gemm_2048x2048x2048_64x64x64_2_cols_0_bcolmaj_1_ccolmaj_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.99 (+1.69%)0.97 (+0.29%)0.97 (-0.04%)0.97 (-0.15%)0.01 (+434.92%)26025.30 (+0.15%)25839.34 (-0.29%)25916.70 (+0.04%)25446.00 (-1.67%)202.05 (+425.44%)
0a6c11c — 2025-12-04 00:39:100.97 (n/a)0.97 (n/a)0.97 (n/a)0.97 (n/a)0.00 (n/a)25986.90 (n/a)25913.52 (n/a)25905.70 (n/a)25877.00 (n/a)38.45 (n/a)

gemm_2048x2048x2048_64x64x64_2_cols_0_bcolmaj_1_ccolmaj_0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
a4b6ffe — 2026-01-05 19:24:050.97 (-0.01%)0.97 (+0.03%)0.97 (+0.04%)0.97 (+0.00%)0.00 (-6.43%)25927.90 (-0.00%)25903.20 (-0.03%)25898.60 (-0.04%)25892.80 (+0.01%)14.05 (-6.56%)663.50 (-0.01%)663.23 (+0.03%)663.35 (+0.04%)662.60 (+0.00%)0.36 (-6.43%)
50ad37c — 2025-12-29 20:35:530.97 (n/a)0.97 (n/a)0.97 (n/a)0.97 (n/a)0.00 (n/a)25928.10 (n/a)25910.14 (n/a)25908.10 (n/a)25891.20 (n/a)15.04 (n/a)663.54 (n/a)663.06 (n/a)663.11 (n/a)662.60 (n/a)0.38 (n/a)

gemm_2048x2048x2048_64x64x64_2_cols_1_bcolmaj_0_ccolmaj_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.92 (-1.02%)0.92 (-0.66%)0.92 (-0.10%)0.91 (-0.78%)0.01 (-4.54%)27751.40 (+0.79%)27480.74 (+0.66%)27354.80 (+0.10%)27320.50 (+1.03%)182.69 (-2.84%)
0a6c11c — 2025-12-04 00:39:100.93 (n/a)0.92 (n/a)0.92 (n/a)0.91 (n/a)0.01 (n/a)27534.80 (n/a)27299.32 (n/a)27326.70 (n/a)27042.30 (n/a)188.04 (n/a)

gemm_2048x2048x2048_64x64x64_2_cols_1_bcolmaj_0_ccolmaj_0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
a4b6ffe — 2026-01-05 19:24:050.92 (+0.11%)0.91 (-0.12%)0.91 (-0.21%)0.91 (-0.24%)0.00 (+30.34%)27723.50 (+0.24%)27548.94 (+0.12%)27522.90 (+0.21%)27403.60 (-0.11%)124.26 (+30.58%)626.92 (+0.11%)623.62 (-0.12%)624.20 (-0.21%)619.69 (-0.24%)2.81 (+30.34%)
50ad37c — 2025-12-29 20:35:530.92 (n/a)0.91 (n/a)0.92 (n/a)0.91 (n/a)0.00 (n/a)27657.10 (n/a)27515.58 (n/a)27464.70 (n/a)27434.60 (n/a)95.16 (n/a)626.21 (n/a)624.37 (n/a)625.52 (n/a)621.17 (n/a)2.16 (n/a)

gemm_2048x2048x2048_64x64x64_2cols_bcolmaj

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
02b4f01 — 2026-02-07 00:16:310.92 (-0.65%)0.91 (-0.66%)0.91 (-0.84%)0.89 (-1.80%)0.01 (+50.92%)28338.00 (+1.84%)27718.36 (+0.67%)27752.40 (+0.84%)27282.00 (+0.65%)421.44 (+54.74%)629.71 (-0.65%)619.92 (-0.66%)619.04 (-0.84%)606.25 (-1.80%)9.37 (+50.92%)
1531781 — 2026-01-23 18:49:240.93 (n/a)0.91 (n/a)0.91 (n/a)0.90 (n/a)0.01 (n/a)27826.70 (n/a)27533.84 (n/a)27519.90 (n/a)27105.20 (n/a)272.35 (n/a)633.82 (n/a)624.00 (n/a)624.27 (n/a)617.39 (n/a)6.21 (n/a)

gemm_2048x2048x2048_64x64x64_8_cols_0_bcolmaj_0_ccolmaj_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)GFLOP/s (max)GFLOP/s (mean)GFLOP/s (median)GFLOP/s (min)GFLOP/s (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
31ab24e — 2025-11-22 02:01:563.65 (-1.06%)3.59 (+0.83%)3.63 (-0.26%)3.45 (+2.66%)0.07 (-44.71%)2678.08 (-1.06%)2632.13 (+0.83%)2658.60 (-0.26%)2528.31 (+2.66%)53.86 (-44.71%)6795.00 (-2.59%)6529.80 (-0.92%)6462.00 (+0.26%)6415.00 (+1.07%)137.14 (-45.27%)
d9a64c5 — 2025-11-21 20:49:353.69 (n/a)3.56 (n/a)3.64 (n/a)3.36 (n/a)0.13 (n/a)2706.77 (n/a)2610.50 (n/a)2665.61 (n/a)2462.71 (n/a)97.41 (n/a)6976.00 (n/a)6590.40 (n/a)6445.00 (n/a)6347.00 (n/a)250.58 (n/a)

gemm_2048x2048x2048_64x64x64_8_cols_0_bcolmaj_0_ccolmaj_0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
d48746f — 2026-01-12 22:17:103.98 (+0.20%)3.84 (-2.95%)3.76 (-5.27%)3.72 (-4.97%)0.13 (+473.97%)6760.00 (+5.23%)6559.50 (+3.13%)6697.10 (+5.56%)6321.00 (-0.20%)218.68 (+499.40%)2717.91 (+0.20%)2621.44 (-2.95%)2565.27 (-5.27%)2541.40 (-4.97%)88.37 (+473.97%)
331dcca — 2026-01-08 18:34:103.97 (n/a)3.96 (n/a)3.97 (n/a)3.92 (n/a)0.02 (n/a)6423.80 (n/a)6360.32 (n/a)6344.30 (n/a)6333.90 (n/a)36.48 (n/a)2712.39 (n/a)2701.17 (n/a)2707.91 (n/a)2674.40 (n/a)15.40 (n/a)

gemm_2048x2048x2048_64x64x64_8_cols_0_bcolmaj_1_ccolmaj_0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
d48746f — 2026-01-12 22:17:103.75 (+0.28%)3.70 (+1.25%)3.71 (-0.21%)3.65 (+4.77%)0.04 (-66.90%)6901.80 (-4.56%)6795.46 (-1.30%)6790.20 (+0.21%)6715.90 (-0.28%)67.05 (-68.40%)2558.08 (+0.28%)2528.33 (+1.25%)2530.10 (-0.21%)2489.18 (+4.77%)24.81 (-66.90%)
331dcca — 2026-01-08 18:34:103.74 (n/a)3.66 (n/a)3.71 (n/a)3.48 (n/a)0.11 (n/a)7231.20 (n/a)6885.18 (n/a)6775.70 (n/a)6734.90 (n/a)212.21 (n/a)2550.87 (n/a)2497.04 (n/a)2535.50 (n/a)2375.78 (n/a)74.96 (n/a)

gemm_2048x2048x2048_64x64x64_8_cols_1_bcolmaj_0_ccolmaj_0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
d48746f — 2026-01-12 22:17:103.62 (-0.77%)3.54 (+0.14%)3.60 (-0.04%)3.44 (+2.17%)0.09 (-25.77%)7307.40 (-2.13%)7103.46 (-0.18%)6982.50 (+0.04%)6948.20 (+0.77%)182.95 (-26.55%)2472.55 (-0.77%)2419.80 (+0.14%)2460.41 (-0.04%)2351.04 (+2.17%)61.76 (-25.77%)
331dcca — 2026-01-08 18:34:103.65 (n/a)3.54 (n/a)3.61 (n/a)3.37 (n/a)0.12 (n/a)7466.20 (n/a)7116.40 (n/a)6979.60 (n/a)6895.10 (n/a)249.09 (n/a)2491.61 (n/a)2416.45 (n/a)2461.45 (n/a)2301.00 (n/a)83.20 (n/a)

gemm_2048x2048x2048_64x64x64_8cols_bcolmaj_ccolmaj

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
02b4f01 — 2026-02-07 00:16:313.36 (-3.14%)3.34 (-2.39%)3.33 (-3.32%)3.32 (+0.81%)0.01 (-79.84%)7589.60 (-0.81%)7546.08 (+2.41%)7551.40 (+3.43%)7499.00 (+3.24%)33.48 (-79.40%)2290.96 (-3.14%)2276.69 (-2.39%)2275.05 (-3.32%)2263.61 (+0.81%)10.11 (-79.84%)
1531781 — 2026-01-23 18:49:243.46 (n/a)3.42 (n/a)3.45 (n/a)3.29 (n/a)0.07 (n/a)7651.20 (n/a)7368.34 (n/a)7300.80 (n/a)7263.70 (n/a)162.53 (n/a)2365.17 (n/a)2332.46 (n/a)2353.14 (n/a)2245.39 (n/a)50.15 (n/a)

gemm_384x1536x1792_32x48x64_4cols_bcolmaj

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
02b4f01 — 2026-02-07 00:16:314.23 (-2.90%)3.79 (-10.63%)4.08 (-5.64%)2.95 (-23.45%)0.56 (+155.63%)2736.60 (+30.63%)2169.64 (+13.85%)1973.40 (+5.98%)1907.30 (+2.99%)358.38 (+238.46%)1108.31 (-2.90%)993.67 (-10.63%)1071.22 (-5.64%)772.47 (-23.45%)147.02 (+155.64%)
1531781 — 2026-01-23 18:49:244.35 (n/a)4.24 (n/a)4.33 (n/a)3.85 (n/a)0.22 (n/a)2094.90 (n/a)1905.64 (n/a)1862.00 (n/a)1852.00 (n/a)105.88 (n/a)1141.45 (n/a)1111.86 (n/a)1135.29 (n/a)1009.08 (n/a)57.51 (n/a)

gemm_896x1792x640_32x64x80_8cols_ccolmaj

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
02b4f01 — 2026-02-07 00:16:316.54 (-3.86%)5.35 (-16.82%)4.93 (-26.43%)4.51 (-15.43%)0.93 (+51.36%)1474.60 (+18.24%)1272.38 (+22.03%)1349.70 (+35.94%)1017.60 (+4.02%)210.81 (+83.97%)2019.72 (-3.86%)1653.44 (-16.82%)1522.73 (-26.43%)1393.72 (-15.43%)288.84 (+51.36%)
1531781 — 2026-01-23 18:49:246.80 (n/a)6.43 (n/a)6.70 (n/a)5.33 (n/a)0.62 (n/a)1247.10 (n/a)1042.70 (n/a)992.90 (n/a)978.30 (n/a)114.59 (n/a)2100.72 (n/a)1987.76 (n/a)2069.85 (n/a)1647.93 (n/a)190.83 (n/a)

layer_norm_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.07 (-36.06%)0.06 (-36.92%)0.05 (-39.76%)0.04 (-36.43%)0.01 (-40.47%)188.40 (+57.39%)153.50 (+57.53%)166.90 (+66.07%)111.80 (+56.36%)32.61 (+43.84%)
1531781 — 2026-01-23 18:49:240.11 (n/a)0.09 (n/a)0.08 (n/a)0.07 (n/a)0.02 (n/a)119.70 (n/a)97.44 (n/a)100.50 (n/a)71.50 (n/a)22.67 (n/a)

layer_norm_1_cols_2_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.07 (-34.19%)0.05 (-43.80%)0.04 (-47.04%)0.04 (-35.62%)0.01 (-35.60%)214.60 (+55.39%)175.36 (+77.53%)183.30 (+88.77%)116.10 (+51.96%)36.46 (+46.56%)
1531781 — 2026-01-23 18:49:240.11 (n/a)0.09 (n/a)0.08 (n/a)0.06 (n/a)0.02 (n/a)138.10 (n/a)98.78 (n/a)97.10 (n/a)76.40 (n/a)24.88 (n/a)

layer_norm_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.08 (-30.19%)0.06 (-31.19%)0.05 (-38.24%)0.04 (-23.76%)0.02 (-24.37%)183.80 (+31.19%)143.52 (+45.88%)152.80 (+61.86%)97.50 (+43.17%)36.31 (+39.09%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.09 (n/a)0.09 (n/a)0.06 (n/a)0.02 (n/a)140.10 (n/a)98.38 (n/a)94.40 (n/a)68.10 (n/a)26.11 (n/a)

layer_norm_2_cols_2_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.07 (-49.98%)0.05 (-51.95%)0.04 (-54.12%)0.03 (-46.32%)0.01 (-51.04%)242.90 (+86.27%)189.94 (+106.68%)203.80 (+117.97%)124.00 (+100.00%)45.52 (+77.88%)
1531781 — 2026-01-23 18:49:240.13 (n/a)0.09 (n/a)0.09 (n/a)0.06 (n/a)0.03 (n/a)130.40 (n/a)91.90 (n/a)93.50 (n/a)62.00 (n/a)25.59 (n/a)

layer_norm_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.05 (-58.37%)0.04 (-56.36%)0.05 (-56.50%)0.03 (-54.09%)0.01 (-61.88%)307.10 (+117.80%)206.96 (+126.09%)179.10 (+129.91%)175.50 (+140.41%)56.45 (+98.60%)
1531781 — 2026-01-23 18:49:240.11 (n/a)0.09 (n/a)0.11 (n/a)0.06 (n/a)0.02 (n/a)141.00 (n/a)91.54 (n/a)77.90 (n/a)73.00 (n/a)28.43 (n/a)

layer_norm_4_cols_2_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.05 (-55.24%)0.04 (-51.44%)0.04 (-52.37%)0.03 (-52.44%)0.01 (-61.58%)312.10 (+110.31%)208.18 (+101.84%)184.30 (+109.91%)164.70 (+123.47%)60.03 (+85.88%)
1531781 — 2026-01-23 18:49:240.11 (n/a)0.09 (n/a)0.09 (n/a)0.06 (n/a)0.02 (n/a)148.40 (n/a)103.14 (n/a)87.80 (n/a)73.70 (n/a)32.30 (n/a)

layer_norm_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.06 (-52.24%)0.05 (-52.19%)0.05 (-55.16%)0.04 (-27.79%)0.01 (-72.58%)213.10 (+38.47%)176.28 (+95.09%)172.60 (+123.00%)140.90 (+109.36%)26.77 (-25.37%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.10 (n/a)0.11 (n/a)0.05 (n/a)0.03 (n/a)153.90 (n/a)90.36 (n/a)77.40 (n/a)67.30 (n/a)35.86 (n/a)

layer_norm_8_cols_2_channels_2048_tile_128

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.05 (-48.26%)0.04 (-54.19%)0.04 (-53.88%)0.02 (-67.70%)0.01 (-14.59%)434.40 (+209.62%)251.20 (+135.87%)211.70 (+116.91%)178.60 (+93.29%)105.02 (+429.42%)
1531781 — 2026-01-23 18:49:240.09 (n/a)0.08 (n/a)0.08 (n/a)0.06 (n/a)0.01 (n/a)140.30 (n/a)106.50 (n/a)97.60 (n/a)92.40 (n/a)19.84 (n/a)

matrix_vector_mul_128x128_32_1col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:490.37 (+17.13%)0.30 (+24.87%)0.30 (+38.03%)0.23 (+12.47%)0.06 (+35.23%)
0a6c11c — 2025-12-04 00:39:100.32 (n/a)0.24 (n/a)0.22 (n/a)0.20 (n/a)0.04 (n/a)

matrix_vector_mul_128x128_32_1col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:100.26 (-1.81%)0.24 (+4.06%)0.24 (+0.52%)0.21 (+16.97%)0.02 (-48.16%)0.26 (-1.81%)0.24 (+4.06%)0.24 (+0.52%)0.20 (+16.97%)0.02 (-48.16%)
a4b6ffe — 2026-01-05 19:24:050.26 (n/a)0.23 (n/a)0.24 (n/a)0.18 (n/a)0.04 (n/a)0.26 (n/a)0.23 (n/a)0.24 (n/a)0.17 (n/a)0.04 (n/a)

matrix_vector_mul_128x128_32tsi_128tso_1col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
02b4f01 — 2026-02-07 00:16:310.28 (-37.89%)0.24 (-30.22%)0.25 (-27.39%)0.19 (-15.10%)0.03 (-58.73%)0.27 (-37.89%)0.24 (-30.22%)0.25 (-27.39%)0.18 (-15.10%)0.03 (-58.73%)
1531781 — 2026-01-23 18:49:240.45 (n/a)0.35 (n/a)0.35 (n/a)0.22 (n/a)0.08 (n/a)0.44 (n/a)0.34 (n/a)0.34 (n/a)0.22 (n/a)0.08 (n/a)

matrix_vector_mul_2048x8192_1_1col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:4913.57 (+3.47%)13.15 (+1.09%)13.10 (+0.16%)12.52 (-1.58%)0.37 (+153.19%)
0a6c11c — 2025-12-04 00:39:1013.11 (n/a)13.01 (n/a)13.07 (n/a)12.72 (n/a)0.15 (n/a)

matrix_vector_mul_2048x8192_1_1col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:1013.43 (-0.07%)13.30 (+4.62%)13.27 (+1.29%)13.16 (+16.70%)0.12 (-86.69%)13.42 (-0.07%)13.29 (+4.62%)13.26 (+1.29%)13.16 (+16.70%)0.12 (-86.69%)
a4b6ffe — 2026-01-05 19:24:0513.44 (n/a)12.71 (n/a)13.10 (n/a)11.28 (n/a)0.87 (n/a)13.43 (n/a)12.70 (n/a)13.09 (n/a)11.27 (n/a)0.87 (n/a)

matrix_vector_mul_2048x8192_1_2col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:4925.06 (+2.47%)23.71 (+1.81%)24.13 (+0.50%)20.90 (+5.41%)1.48 (-15.60%)
0a6c11c — 2025-12-04 00:39:1024.46 (n/a)23.29 (n/a)24.01 (n/a)19.83 (n/a)1.75 (n/a)

matrix_vector_mul_2048x8192_1_2col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:1024.46 (-0.60%)21.07 (-8.03%)19.75 (-17.83%)18.08 (+1.18%)3.03 (+7.07%)24.44 (-0.60%)21.06 (-8.03%)19.74 (-17.83%)18.07 (+1.18%)3.03 (+7.07%)
a4b6ffe — 2026-01-05 19:24:0524.60 (n/a)22.91 (n/a)24.04 (n/a)17.87 (n/a)2.83 (n/a)24.59 (n/a)22.90 (n/a)24.02 (n/a)17.86 (n/a)2.83 (n/a)

matrix_vector_mul_2048x8192_1_4col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:4942.05 (+1.54%)40.50 (+12.60%)40.49 (+2.36%)38.98 (+97.53%)0.97 (-88.09%)
0a6c11c — 2025-12-04 00:39:1041.41 (n/a)35.97 (n/a)39.55 (n/a)19.73 (n/a)8.15 (n/a)

matrix_vector_mul_2048x8192_1_4col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:1042.58 (+1.51%)41.37 (+14.29%)41.25 (+4.89%)40.33 (+91.42%)0.92 (-89.18%)42.56 (+1.51%)41.34 (+14.29%)41.23 (+4.89%)40.30 (+91.42%)0.92 (-89.18%)
a4b6ffe — 2026-01-05 19:24:0541.95 (n/a)36.20 (n/a)39.33 (n/a)21.07 (n/a)8.54 (n/a)41.92 (n/a)36.17 (n/a)39.31 (n/a)21.05 (n/a)8.53 (n/a)

matrix_vector_mul_2048x8192_1_8col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:4946.62 (+3.11%)44.82 (+0.17%)44.78 (+0.17%)42.60 (-3.80%)1.60 (+367.72%)
0a6c11c — 2025-12-04 00:39:1045.21 (n/a)44.75 (n/a)44.70 (n/a)44.29 (n/a)0.34 (n/a)

matrix_vector_mul_2048x8192_1_8col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:1046.04 (+4.14%)45.06 (+3.47%)44.55 (+2.76%)44.30 (+2.84%)0.87 (+66.58%)46.01 (+4.14%)45.03 (+3.47%)44.52 (+2.76%)44.27 (+2.84%)0.87 (+66.58%)
a4b6ffe — 2026-01-05 19:24:0544.21 (n/a)43.55 (n/a)43.35 (n/a)43.07 (n/a)0.52 (n/a)44.18 (n/a)43.52 (n/a)43.32 (n/a)43.05 (n/a)0.52 (n/a)

matrix_vector_mul_2048x8192_1tsi_1024tso_2col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
02b4f01 — 2026-02-07 00:16:3124.92 (+0.45%)24.16 (-1.77%)24.11 (-1.76%)23.64 (-3.10%)0.52 (+196.63%)24.90 (+0.45%)24.14 (-1.77%)24.10 (-1.76%)23.63 (-3.10%)0.52 (+196.63%)
1531781 — 2026-01-23 18:49:2424.80 (n/a)24.59 (n/a)24.55 (n/a)24.40 (n/a)0.18 (n/a)24.79 (n/a)24.58 (n/a)24.53 (n/a)24.38 (n/a)0.18 (n/a)

matrix_vector_mul_2048x8192_1tsi_2048tso_1col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
02b4f01 — 2026-02-07 00:16:3113.41 (-0.97%)11.94 (-10.52%)12.08 (-10.19%)9.21 (-29.73%)1.67 (+708.22%)13.40 (-0.97%)11.93 (-10.52%)12.07 (-10.19%)9.20 (-29.73%)1.67 (+708.22%)
1531781 — 2026-01-23 18:49:2413.54 (n/a)13.34 (n/a)13.45 (n/a)13.10 (n/a)0.21 (n/a)13.53 (n/a)13.33 (n/a)13.44 (n/a)13.09 (n/a)0.21 (n/a)

matrix_vector_mul_2048x8192_1tsi_256tso_8col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
02b4f01 — 2026-02-07 00:16:3144.96 (-3.95%)43.37 (-5.54%)42.92 (-7.45%)42.33 (-5.55%)1.20 (+34.77%)44.94 (-3.95%)43.34 (-5.54%)42.89 (-7.45%)42.30 (-5.55%)1.20 (+34.77%)
1531781 — 2026-01-23 18:49:2446.81 (n/a)45.91 (n/a)46.38 (n/a)44.81 (n/a)0.89 (n/a)46.79 (n/a)45.88 (n/a)46.35 (n/a)44.79 (n/a)0.89 (n/a)

matrix_vector_mul_2048x8192_1tsi_512tso_4col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
02b4f01 — 2026-02-07 00:16:3140.75 (-1.79%)39.88 (-2.29%)39.64 (-3.62%)39.47 (-1.21%)0.53 (-19.23%)40.72 (-1.79%)39.85 (-2.29%)39.61 (-3.62%)39.44 (-1.21%)0.53 (-19.23%)
1531781 — 2026-01-23 18:49:2441.49 (n/a)40.81 (n/a)41.13 (n/a)39.95 (n/a)0.65 (n/a)41.46 (n/a)40.79 (n/a)41.10 (n/a)39.93 (n/a)0.65 (n/a)

matrix_vector_mul_8192x2048_4_1col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:4913.29 (+1.63%)13.19 (+6.08%)13.19 (+1.17%)13.09 (+26.86%)0.08 (-92.94%)
0a6c11c — 2025-12-04 00:39:1013.08 (n/a)12.44 (n/a)13.03 (n/a)10.32 (n/a)1.07 (n/a)

matrix_vector_mul_8192x2048_4_1col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:1013.68 (+3.42%)13.30 (+8.54%)13.38 (+7.25%)12.72 (+14.96%)0.35 (-66.09%)13.67 (+3.42%)13.29 (+8.54%)13.38 (+7.25%)12.71 (+14.96%)0.35 (-66.09%)
a4b6ffe — 2026-01-05 19:24:0513.23 (n/a)12.25 (n/a)12.48 (n/a)11.07 (n/a)1.04 (n/a)13.22 (n/a)12.24 (n/a)12.47 (n/a)11.06 (n/a)1.04 (n/a)

matrix_vector_mul_8192x2048_4_2col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:4925.58 (+4.24%)24.23 (+6.59%)24.03 (+0.29%)23.58 (+37.29%)0.70 (-74.97%)
0a6c11c — 2025-12-04 00:39:1024.54 (n/a)22.73 (n/a)23.96 (n/a)17.18 (n/a)2.79 (n/a)

matrix_vector_mul_8192x2048_4_2col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:1025.08 (+1.32%)24.68 (+13.42%)24.61 (+3.26%)24.47 (+48.32%)0.24 (-93.56%)25.07 (+1.32%)24.67 (+13.42%)24.59 (+3.26%)24.46 (+48.32%)0.24 (-93.56%)
a4b6ffe — 2026-01-05 19:24:0524.76 (n/a)21.76 (n/a)23.83 (n/a)16.50 (n/a)3.74 (n/a)24.74 (n/a)21.75 (n/a)23.82 (n/a)16.49 (n/a)3.73 (n/a)

matrix_vector_mul_8192x2048_4_4col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:4940.94 (-2.19%)39.74 (-1.10%)39.76 (-0.60%)38.17 (-2.14%)1.04 (+4.39%)
0a6c11c — 2025-12-04 00:39:1041.86 (n/a)40.18 (n/a)40.00 (n/a)39.00 (n/a)1.00 (n/a)

matrix_vector_mul_8192x2048_4_4col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:1040.74 (-1.09%)37.16 (-7.15%)40.25 (+1.47%)24.51 (-37.44%)7.07 (+736.13%)40.72 (-1.09%)37.13 (-7.15%)40.23 (+1.47%)24.50 (-37.44%)7.07 (+736.13%)
a4b6ffe — 2026-01-05 19:24:0541.19 (n/a)40.02 (n/a)39.67 (n/a)39.18 (n/a)0.85 (n/a)41.17 (n/a)39.99 (n/a)39.64 (n/a)39.16 (n/a)0.85 (n/a)

matrix_vector_mul_8192x2048_4_8col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:4946.71 (-1.04%)44.91 (-2.34%)44.71 (-2.98%)43.80 (-1.44%)0.96 (+6.93%)
0a6c11c — 2025-12-04 00:39:1047.20 (n/a)45.98 (n/a)46.08 (n/a)44.44 (n/a)0.90 (n/a)

matrix_vector_mul_8192x2048_4_8col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:1046.28 (+5.68%)44.89 (+14.59%)44.74 (+4.26%)44.20 (+62.80%)0.84 (-87.96%)46.25 (+5.68%)44.86 (+14.59%)44.71 (+4.26%)44.17 (+62.80%)0.84 (-87.96%)
a4b6ffe — 2026-01-05 19:24:0543.79 (n/a)39.17 (n/a)42.91 (n/a)27.15 (n/a)6.98 (n/a)43.77 (n/a)39.15 (n/a)42.88 (n/a)27.13 (n/a)6.98 (n/a)

matrix_vector_mul_8192x2048_4tsi_1024tso_1col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
02b4f01 — 2026-02-07 00:16:3113.43 (-0.19%)13.21 (-0.85%)13.21 (-0.81%)13.08 (-0.93%)0.14 (+22.62%)13.43 (-0.19%)13.20 (-0.85%)13.20 (-0.81%)13.07 (-0.93%)0.14 (+22.62%)
1531781 — 2026-01-23 18:49:2413.46 (n/a)13.33 (n/a)13.32 (n/a)13.20 (n/a)0.11 (n/a)13.45 (n/a)13.32 (n/a)13.31 (n/a)13.19 (n/a)0.11 (n/a)

matrix_vector_mul_8192x2048_4tsi_1024tso_2col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
02b4f01 — 2026-02-07 00:16:3124.76 (-2.00%)24.01 (-2.98%)23.93 (-2.78%)23.69 (-3.50%)0.43 (+45.53%)24.75 (-2.00%)24.00 (-2.98%)23.91 (-2.78%)23.67 (-3.50%)0.43 (+45.53%)
1531781 — 2026-01-23 18:49:2425.27 (n/a)24.75 (n/a)24.61 (n/a)24.55 (n/a)0.30 (n/a)25.25 (n/a)24.74 (n/a)24.59 (n/a)24.53 (n/a)0.30 (n/a)

matrix_vector_mul_8192x2048_4tsi_1024tso_4col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
02b4f01 — 2026-02-07 00:16:3141.37 (+1.24%)39.60 (-1.02%)39.12 (-3.25%)38.31 (-1.43%)1.30 (+46.28%)41.34 (+1.24%)39.57 (-1.02%)39.09 (-3.25%)38.29 (-1.43%)1.30 (+46.28%)
1531781 — 2026-01-23 18:49:2440.86 (n/a)40.01 (n/a)40.43 (n/a)38.86 (n/a)0.89 (n/a)40.84 (n/a)39.98 (n/a)40.41 (n/a)38.84 (n/a)0.89 (n/a)

matrix_vector_mul_8192x2048_4tsi_1024tso_8col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
02b4f01 — 2026-02-07 00:16:3143.45 (-5.71%)42.88 (-4.94%)42.80 (-5.91%)42.35 (-3.51%)0.47 (-51.32%)43.42 (-5.71%)42.85 (-4.94%)42.78 (-5.91%)42.33 (-3.51%)0.47 (-51.32%)
1531781 — 2026-01-23 18:49:2446.08 (n/a)45.11 (n/a)45.49 (n/a)43.89 (n/a)0.97 (n/a)46.05 (n/a)45.08 (n/a)45.46 (n/a)43.87 (n/a)0.97 (n/a)

mem_copy_16_cores_2_chans_2048_tile_128_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.07 (+22.03%)0.06 (+27.86%)0.06 (+35.49%)0.05 (+15.00%)0.01 (+49.69%)173.30 (-13.00%)137.94 (-21.32%)133.60 (-26.19%)123.40 (-18.06%)18.40 (+8.53%)
0a6c11c — 2025-12-04 00:39:100.05 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.00 (n/a)199.20 (n/a)175.32 (n/a)181.00 (n/a)150.60 (n/a)16.96 (n/a)

mem_copy_16_cores_2_chans_2048_tile_128_False0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.05 (-53.48%)0.04 (-53.01%)0.04 (-59.95%)0.03 (-28.20%)0.01 (-76.42%)272.00 (+39.27%)218.28 (+95.56%)216.20 (+149.65%)181.90 (+115.01%)33.90 (-28.86%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.08 (n/a)0.09 (n/a)0.04 (n/a)0.02 (n/a)195.30 (n/a)111.62 (n/a)86.60 (n/a)84.60 (n/a)47.65 (n/a)

mem_copy_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
003b49d — 2025-11-14 19:30:540.05 (-16.99%)0.05 (-12.47%)0.05 (-7.32%)0.04 (-11.17%)0.00 (-29.61%)197.00 (+12.57%)171.00 (+13.70%)164.00 (+7.89%)153.00 (+20.47%)17.49 (-4.32%)
0999a9a — 2025-11-13 19:24:080.06 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)175.00 (n/a)150.40 (n/a)152.00 (n/a)127.00 (n/a)18.28 (n/a)

mem_copy_1_cols_2_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
003b49d — 2025-11-14 19:30:540.06 (+5.56%)0.05 (-1.59%)0.05 (-1.39%)0.03 (-14.07%)0.01 (+25.03%)263.00 (+16.37%)166.00 (+4.53%)144.00 (+1.41%)126.00 (-5.26%)49.38 (+43.17%)
0999a9a — 2025-11-13 19:24:080.06 (n/a)0.05 (n/a)0.05 (n/a)0.03 (n/a)0.01 (n/a)226.00 (n/a)158.80 (n/a)142.00 (n/a)133.00 (n/a)34.49 (n/a)

mem_copy_1_cores_1_chans_2048_tile_2048_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.08 (+10.93%)0.07 (+7.79%)0.07 (+3.55%)0.05 (-1.73%)0.01 (+36.73%)150.20 (+1.76%)121.68 (-6.36%)125.60 (-3.38%)97.90 (-9.85%)18.56 (+21.98%)
0a6c11c — 2025-12-04 00:39:100.08 (n/a)0.06 (n/a)0.06 (n/a)0.06 (n/a)0.01 (n/a)147.60 (n/a)129.94 (n/a)130.00 (n/a)108.60 (n/a)15.21 (n/a)

mem_copy_1_cores_1_chans_2048_tile_2048_False0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.07 (-28.32%)0.05 (-39.34%)0.05 (-50.54%)0.05 (-33.59%)0.01 (-31.15%)177.90 (+50.64%)159.08 (+64.82%)170.70 (+102.25%)114.90 (+39.44%)26.34 (+44.02%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.09 (n/a)0.10 (n/a)0.07 (n/a)0.02 (n/a)118.10 (n/a)96.52 (n/a)84.40 (n/a)82.40 (n/a)18.29 (n/a)

mem_copy_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
003b49d — 2025-11-14 19:30:540.06 (-15.32%)0.05 (-17.09%)0.05 (-23.81%)0.04 (-15.54%)0.01 (-18.18%)193.00 (+18.40%)160.80 (+20.36%)168.00 (+31.25%)124.00 (+18.10%)22.92 (+11.39%)
0999a9a — 2025-11-13 19:24:080.07 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)163.00 (n/a)133.60 (n/a)128.00 (n/a)105.00 (n/a)20.58 (n/a)

mem_copy_2_cols_2_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
003b49d — 2025-11-14 19:30:540.06 (-23.02%)0.05 (-6.98%)0.05 (-2.13%)0.04 (-15.31%)0.01 (-33.29%)196.00 (+18.07%)147.00 (+6.52%)141.00 (+2.17%)126.00 (+29.90%)25.27 (+8.13%)
0999a9a — 2025-11-13 19:24:080.08 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)166.00 (n/a)138.00 (n/a)138.00 (n/a)97.00 (n/a)23.37 (n/a)

mem_copy_2_cores_1_chans_2048_tile_1024_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.10 (+31.37%)0.08 (+29.12%)0.08 (+29.06%)0.07 (+41.80%)0.01 (+0.23%)109.90 (-29.46%)98.30 (-23.38%)100.80 (-22.52%)79.40 (-23.87%)10.54 (-46.17%)
0a6c11c — 2025-12-04 00:39:100.08 (n/a)0.07 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)155.80 (n/a)128.30 (n/a)130.10 (n/a)104.30 (n/a)19.58 (n/a)

mem_copy_2_cores_1_chans_2048_tile_1024_False0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.06 (-50.57%)0.05 (-45.81%)0.05 (-46.88%)0.04 (-39.27%)0.01 (-70.88%)192.90 (+64.73%)164.04 (+77.92%)163.60 (+88.26%)142.30 (+102.13%)22.07 (-5.36%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.09 (n/a)0.09 (n/a)0.07 (n/a)0.02 (n/a)117.10 (n/a)92.20 (n/a)86.90 (n/a)70.40 (n/a)23.32 (n/a)

mem_copy_2_cores_2_chans_2048_tile_1024_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.10 (+25.02%)0.07 (+10.01%)0.07 (+7.53%)0.05 (-1.99%)0.01 (+77.08%)150.10 (+2.04%)114.60 (-7.22%)111.70 (-6.99%)83.90 (-20.02%)21.23 (+43.54%)
0a6c11c — 2025-12-04 00:39:100.08 (n/a)0.07 (n/a)0.07 (n/a)0.06 (n/a)0.01 (n/a)147.10 (n/a)123.52 (n/a)120.10 (n/a)104.90 (n/a)14.79 (n/a)

mem_copy_2_cores_2_chans_2048_tile_1024_False0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.05 (-45.30%)0.05 (-43.46%)0.05 (-40.96%)0.04 (-37.85%)0.01 (-53.89%)225.50 (+60.84%)177.16 (+74.27%)155.90 (+69.27%)151.60 (+82.87%)32.88 (+35.72%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.08 (n/a)0.09 (n/a)0.06 (n/a)0.02 (n/a)140.20 (n/a)101.66 (n/a)92.10 (n/a)82.90 (n/a)24.23 (n/a)

mem_copy_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
003b49d — 2025-11-14 19:30:540.07 (+11.97%)0.05 (+3.09%)0.05 (+2.10%)0.03 (-11.35%)0.01 (+41.65%)229.00 (+12.81%)156.00 (-0.51%)143.00 (-2.05%)117.00 (-10.69%)38.21 (+48.71%)
0999a9a — 2025-11-13 19:24:080.06 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)203.00 (n/a)156.80 (n/a)146.00 (n/a)131.00 (n/a)25.69 (n/a)

mem_copy_4_cols_2_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
003b49d — 2025-11-14 19:30:540.06 (+9.02%)0.05 (+5.59%)0.05 (+0.00%)0.04 (+3.63%)0.01 (+25.48%)193.00 (-3.50%)147.40 (-4.66%)147.00 (+0.00%)122.00 (-8.27%)25.24 (+7.48%)
0999a9a — 2025-11-13 19:24:080.06 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)200.00 (n/a)154.60 (n/a)147.00 (n/a)133.00 (n/a)23.48 (n/a)

mem_copy_4_cores_1_chans_2048_tile_512_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.09 (-3.98%)0.07 (+4.00%)0.07 (+7.04%)0.06 (+3.67%)0.01 (-15.76%)129.70 (-3.57%)113.30 (-4.50%)116.90 (-6.55%)90.40 (+4.15%)15.04 (-13.80%)
0a6c11c — 2025-12-04 00:39:100.09 (n/a)0.07 (n/a)0.07 (n/a)0.06 (n/a)0.01 (n/a)134.50 (n/a)118.64 (n/a)125.10 (n/a)86.80 (n/a)17.45 (n/a)

mem_copy_4_cores_1_chans_2048_tile_512_False0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.07 (-43.21%)0.05 (-51.23%)0.05 (-58.02%)0.04 (-37.77%)0.01 (-53.70%)206.20 (+60.72%)168.94 (+99.79%)170.50 (+138.13%)115.60 (+75.95%)33.50 (+26.82%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.10 (n/a)0.11 (n/a)0.06 (n/a)0.03 (n/a)128.30 (n/a)84.56 (n/a)71.60 (n/a)65.70 (n/a)26.42 (n/a)

mem_copy_4_cores_2_chans_2048_tile_512_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.08 (+9.57%)0.07 (+16.85%)0.07 (+19.58%)0.06 (+24.52%)0.01 (+1.05%)134.50 (-19.70%)116.64 (-14.92%)116.80 (-16.39%)98.00 (-8.75%)15.71 (-25.13%)
0a6c11c — 2025-12-04 00:39:100.08 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)167.50 (n/a)137.10 (n/a)139.70 (n/a)107.40 (n/a)20.99 (n/a)

mem_copy_4_cores_2_chans_2048_tile_512_False0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.05 (-45.85%)0.05 (-43.38%)0.05 (-46.29%)0.05 (-26.02%)0.00 (-78.92%)180.00 (+35.14%)159.00 (+71.00%)155.20 (+86.09%)151.60 (+84.65%)11.85 (-47.32%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.09 (n/a)0.10 (n/a)0.06 (n/a)0.02 (n/a)133.20 (n/a)92.98 (n/a)83.40 (n/a)82.10 (n/a)22.49 (n/a)

mem_copy_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
003b49d — 2025-11-14 19:30:540.05 (-13.48%)0.05 (-17.79%)0.05 (-15.33%)0.03 (-25.09%)0.01 (+6.67%)267.00 (+33.50%)176.80 (+24.51%)150.00 (+18.11%)141.00 (+15.57%)47.41 (+61.41%)
0999a9a — 2025-11-13 19:24:080.06 (n/a)0.06 (n/a)0.06 (n/a)0.04 (n/a)0.01 (n/a)200.00 (n/a)142.00 (n/a)127.00 (n/a)122.00 (n/a)29.37 (n/a)

mem_copy_8_cols_2_channels_2048_tile_128

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
003b49d — 2025-11-14 19:30:540.05 (+13.38%)0.05 (+23.85%)0.05 (+37.91%)0.05 (+30.72%)0.00 (-56.85%)166.00 (-23.50%)153.40 (-20.60%)153.00 (-27.49%)142.00 (-11.80%)7.63 (-70.60%)
0999a9a — 2025-11-13 19:24:080.05 (n/a)0.04 (n/a)0.04 (n/a)0.04 (n/a)0.01 (n/a)217.00 (n/a)193.20 (n/a)211.00 (n/a)161.00 (n/a)25.96 (n/a)

mem_copy_8_cores_1_chans_2048_tile_256_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.09 (+16.31%)0.08 (+15.34%)0.08 (+14.60%)0.07 (+19.30%)0.01 (-5.41%)117.30 (-16.15%)107.80 (-13.61%)108.50 (-12.78%)94.10 (-13.99%)8.29 (-32.01%)
0a6c11c — 2025-12-04 00:39:100.07 (n/a)0.07 (n/a)0.07 (n/a)0.06 (n/a)0.01 (n/a)139.90 (n/a)124.78 (n/a)124.40 (n/a)109.40 (n/a)12.20 (n/a)

mem_copy_8_cores_1_chans_2048_tile_256_False0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.05 (-44.84%)0.05 (-43.83%)0.05 (-47.91%)0.04 (-38.04%)0.01 (-61.74%)212.00 (+61.46%)175.18 (+75.57%)170.20 (+91.88%)157.60 (+81.36%)21.65 (+14.52%)
1531781 — 2026-01-23 18:49:240.09 (n/a)0.08 (n/a)0.09 (n/a)0.06 (n/a)0.01 (n/a)131.30 (n/a)99.78 (n/a)88.70 (n/a)86.90 (n/a)18.90 (n/a)

mem_copy_8_cores_2_chans_2048_tile_256_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.09 (+12.61%)0.07 (-2.24%)0.07 (-8.85%)0.06 (+5.13%)0.01 (+35.50%)136.50 (-4.88%)120.86 (+2.96%)125.10 (+9.64%)91.60 (-11.15%)15.72 (+10.46%)
0a6c11c — 2025-12-04 00:39:100.08 (n/a)0.07 (n/a)0.07 (n/a)0.06 (n/a)0.01 (n/a)143.50 (n/a)117.38 (n/a)114.10 (n/a)103.10 (n/a)14.23 (n/a)

mem_copy_8_cores_2_chans_2048_tile_256_False0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.08 (-22.85%)0.05 (-39.65%)0.05 (-48.80%)0.04 (-30.69%)0.02 (-12.94%)208.40 (+44.32%)163.84 (+67.70%)172.10 (+95.35%)103.70 (+29.62%)39.18 (+48.73%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.09 (n/a)0.09 (n/a)0.06 (n/a)0.02 (n/a)144.40 (n/a)97.70 (n/a)88.10 (n/a)80.00 (n/a)26.35 (n/a)

mha

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.18 (+0.16%)0.18 (+0.19%)0.18 (+0.12%)0.18 (+0.32%)0.00 (-52.24%)47253.90 (-0.32%)47213.70 (-0.19%)47210.90 (-0.12%)47167.70 (-0.16%)28.71 (-52.47%)
0a6c11c — 2025-12-04 00:39:100.18 (n/a)0.18 (n/a)0.18 (n/a)0.18 (n/a)0.00 (n/a)47405.00 (n/a)47301.72 (n/a)47265.90 (n/a)47241.80 (n/a)60.40 (n/a)

mha0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.18 (-1.08%)0.18 (-0.77%)0.18 (-0.77%)0.18 (-0.51%)0.00 (-91.50%)47609.30 (+0.51%)47597.74 (+0.77%)47596.90 (+0.77%)47584.80 (+1.09%)9.08 (-91.39%)
1531781 — 2026-01-23 18:49:240.18 (n/a)0.18 (n/a)0.18 (n/a)0.18 (n/a)0.00 (n/a)47367.50 (n/a)47232.68 (n/a)47232.00 (n/a)47073.00 (n/a)105.45 (n/a)

relu_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.07 (-41.43%)0.06 (-34.94%)0.06 (-39.65%)0.04 (-21.30%)0.01 (-64.58%)194.30 (+27.08%)146.66 (+44.32%)136.10 (+65.57%)124.70 (+70.82%)28.43 (-20.47%)
1531781 — 2026-01-23 18:49:240.11 (n/a)0.09 (n/a)0.10 (n/a)0.05 (n/a)0.03 (n/a)152.90 (n/a)101.62 (n/a)82.20 (n/a)73.00 (n/a)35.75 (n/a)

relu_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.07 (-31.16%)0.06 (-34.41%)0.05 (-38.90%)0.05 (-13.14%)0.01 (-55.38%)173.30 (+15.07%)149.52 (+47.66%)149.50 (+63.75%)120.60 (+45.30%)20.05 (-28.46%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.08 (n/a)0.09 (n/a)0.05 (n/a)0.02 (n/a)150.60 (n/a)101.26 (n/a)91.30 (n/a)83.00 (n/a)28.02 (n/a)

relu_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.06 (-43.06%)0.05 (-46.13%)0.05 (-49.61%)0.04 (-41.44%)0.01 (-44.92%)232.00 (+70.84%)177.80 (+84.82%)171.20 (+98.38%)145.90 (+75.57%)35.88 (+59.73%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.09 (n/a)0.09 (n/a)0.06 (n/a)0.02 (n/a)135.80 (n/a)96.20 (n/a)86.30 (n/a)83.10 (n/a)22.46 (n/a)

relu_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.05 (-53.79%)0.04 (-57.05%)0.04 (-62.52%)0.03 (-51.31%)0.01 (-57.93%)317.10 (+105.38%)213.32 (+128.59%)209.10 (+166.71%)164.00 (+116.36%)62.33 (+82.04%)
1531781 — 2026-01-23 18:49:240.11 (n/a)0.09 (n/a)0.10 (n/a)0.05 (n/a)0.02 (n/a)154.40 (n/a)93.32 (n/a)78.40 (n/a)75.80 (n/a)34.24 (n/a)

rms_norm_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.07 (-25.86%)0.05 (-34.28%)0.06 (-36.68%)0.04 (-27.99%)0.01 (-24.97%)182.60 (+38.86%)153.32 (+52.25%)145.90 (+57.90%)122.50 (+34.76%)24.41 (+40.71%)
1531781 — 2026-01-23 18:49:240.09 (n/a)0.08 (n/a)0.09 (n/a)0.06 (n/a)0.01 (n/a)131.50 (n/a)100.70 (n/a)92.40 (n/a)90.90 (n/a)17.35 (n/a)

rms_norm_1_cols_2_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.07 (-36.37%)0.06 (-42.10%)0.06 (-47.26%)0.05 (-15.29%)0.01 (-62.37%)171.40 (+18.04%)140.94 (+62.86%)136.60 (+89.46%)110.90 (+57.30%)22.41 (-31.70%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.10 (n/a)0.11 (n/a)0.06 (n/a)0.03 (n/a)145.20 (n/a)86.54 (n/a)72.10 (n/a)70.50 (n/a)32.80 (n/a)

rms_norm_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.07 (-23.91%)0.06 (-31.30%)0.06 (-30.19%)0.04 (-39.06%)0.01 (+4.61%)212.70 (+64.12%)146.86 (+49.52%)130.20 (+43.23%)118.00 (+31.40%)39.42 (+124.61%)
1531781 — 2026-01-23 18:49:240.09 (n/a)0.09 (n/a)0.09 (n/a)0.06 (n/a)0.01 (n/a)129.60 (n/a)98.22 (n/a)90.90 (n/a)89.80 (n/a)17.55 (n/a)

rms_norm_2_cols_2_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.06 (-55.57%)0.04 (-60.96%)0.04 (-60.63%)0.03 (-67.68%)0.01 (-30.10%)289.20 (+209.30%)198.16 (+166.63%)185.70 (+154.04%)140.40 (+125.00%)56.41 (+386.20%)
1531781 — 2026-01-23 18:49:240.13 (n/a)0.11 (n/a)0.11 (n/a)0.09 (n/a)0.02 (n/a)93.50 (n/a)74.32 (n/a)73.10 (n/a)62.40 (n/a)11.60 (n/a)

rms_norm_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.06 (-35.46%)0.05 (-37.36%)0.06 (-36.61%)0.04 (-30.58%)0.01 (-41.95%)188.40 (+44.15%)154.64 (+58.64%)142.30 (+57.76%)133.30 (+55.00%)23.57 (+26.28%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.09 (n/a)0.09 (n/a)0.06 (n/a)0.01 (n/a)130.70 (n/a)97.48 (n/a)90.20 (n/a)86.00 (n/a)18.66 (n/a)

rms_norm_4_cols_2_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.07 (-36.93%)0.05 (-49.99%)0.05 (-55.79%)0.04 (-43.87%)0.01 (-40.39%)223.60 (+78.17%)167.36 (+99.48%)166.20 (+126.12%)115.60 (+58.57%)38.24 (+64.40%)
1531781 — 2026-01-23 18:49:240.11 (n/a)0.10 (n/a)0.11 (n/a)0.07 (n/a)0.02 (n/a)125.50 (n/a)83.90 (n/a)73.50 (n/a)72.90 (n/a)23.26 (n/a)

rms_norm_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.07 (-23.40%)0.04 (-44.97%)0.05 (-45.14%)0.03 (-49.51%)0.02 (+10.67%)305.70 (+98.12%)210.20 (+97.07%)172.90 (+82.38%)122.80 (+30.64%)80.58 (+202.48%)
1531781 — 2026-01-23 18:49:240.09 (n/a)0.08 (n/a)0.09 (n/a)0.05 (n/a)0.01 (n/a)154.30 (n/a)106.66 (n/a)94.80 (n/a)94.00 (n/a)26.64 (n/a)

rms_norm_8_cols_2_channels_2048_tile_128

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.05 (-49.80%)0.04 (-53.94%)0.04 (-56.10%)0.02 (-53.21%)0.01 (-43.58%)358.30 (+113.66%)241.46 (+120.79%)208.60 (+127.73%)177.10 (+99.21%)77.94 (+132.51%)
1531781 — 2026-01-23 18:49:240.09 (n/a)0.08 (n/a)0.09 (n/a)0.05 (n/a)0.02 (n/a)167.70 (n/a)109.36 (n/a)91.60 (n/a)88.90 (n/a)33.52 (n/a)

rope_1_cols_2_channels_4096_tile_4096_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1b2d230 — 2026-01-13 22:38:030.19 (-21.66%)0.17 (-8.53%)0.18 (+2.09%)0.14 (-6.71%)0.02 (-38.07%)174.10 (+7.20%)150.06 (+8.11%)139.60 (-2.04%)129.80 (+27.63%)20.36 (-10.65%)
d48746f — 2026-01-12 22:17:100.24 (n/a)0.18 (n/a)0.17 (n/a)0.15 (n/a)0.04 (n/a)162.40 (n/a)138.80 (n/a)142.50 (n/a)101.70 (n/a)22.79 (n/a)

rope_1c_32rows_512cols_32arows_0m

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.78 (-29.75%)0.62 (-37.45%)0.61 (-42.00%)0.45 (-38.27%)0.14 (-8.16%)216.70 (+61.96%)165.50 (+62.99%)160.90 (+72.45%)126.80 (+42.31%)38.16 (+105.19%)
1531781 — 2026-01-23 18:49:241.10 (n/a)0.99 (n/a)1.05 (n/a)0.73 (n/a)0.15 (n/a)133.80 (n/a)101.54 (n/a)93.30 (n/a)89.10 (n/a)18.60 (n/a)

rope_1c_32rows_512cols_8arows_0m

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.65 (+16.49%)0.55 (+4.23%)0.55 (+5.34%)0.47 (-1.87%)0.07 (+128.83%)156.60 (+1.89%)136.46 (-3.11%)133.00 (-5.07%)113.70 (-14.19%)16.93 (+101.52%)
1531781 — 2026-01-23 18:49:240.56 (n/a)0.52 (n/a)0.53 (n/a)0.48 (n/a)0.03 (n/a)153.70 (n/a)140.84 (n/a)140.10 (n/a)132.50 (n/a)8.40 (n/a)

rope_2_cols_2_channels_4096_tile_2048_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1b2d230 — 2026-01-13 22:38:030.21 (-0.56%)0.17 (-10.62%)0.16 (-13.58%)0.13 (-13.33%)0.03 (+17.05%)183.70 (+15.39%)151.76 (+12.77%)155.20 (+15.73%)117.30 (+0.60%)24.24 (+35.73%)
d48746f — 2026-01-12 22:17:100.21 (n/a)0.19 (n/a)0.18 (n/a)0.15 (n/a)0.02 (n/a)159.20 (n/a)134.58 (n/a)134.10 (n/a)116.60 (n/a)17.86 (n/a)

rope_2c_32rows_512cols_32arows_0m

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.77 (-38.64%)0.59 (-38.86%)0.56 (-38.52%)0.44 (-30.26%)0.13 (-46.19%)225.10 (+43.38%)173.60 (+60.59%)176.20 (+62.70%)128.30 (+63.02%)39.09 (+25.72%)
1531781 — 2026-01-23 18:49:241.25 (n/a)0.97 (n/a)0.91 (n/a)0.63 (n/a)0.25 (n/a)157.00 (n/a)108.10 (n/a)108.30 (n/a)78.70 (n/a)31.09 (n/a)

rope_2c_32rows_512cols_8arows_0m

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.64 (+1.61%)0.44 (-22.42%)0.44 (-22.99%)0.18 (-62.47%)0.19 (+257.67%)401.30 (+166.47%)204.04 (+56.26%)169.30 (+29.83%)114.90 (-1.63%)116.80 (+829.54%)
1531781 — 2026-01-23 18:49:240.63 (n/a)0.57 (n/a)0.57 (n/a)0.49 (n/a)0.05 (n/a)150.60 (n/a)130.58 (n/a)130.40 (n/a)116.80 (n/a)12.57 (n/a)

rope_4_cols_2_channels_4096_tile_1024_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1b2d230 — 2026-01-13 22:38:030.22 (+18.92%)0.17 (+6.24%)0.17 (+5.57%)0.14 (-0.50%)0.03 (+90.77%)171.10 (+0.47%)147.40 (-4.39%)148.10 (-5.31%)110.10 (-15.89%)23.59 (+60.67%)
d48746f — 2026-01-12 22:17:100.19 (n/a)0.16 (n/a)0.16 (n/a)0.14 (n/a)0.02 (n/a)170.30 (n/a)154.16 (n/a)156.40 (n/a)130.90 (n/a)14.69 (n/a)

rope_8_cols_2_channels_4096_tile_512_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1b2d230 — 2026-01-13 22:38:030.17 (-1.30%)0.15 (+6.63%)0.15 (-8.78%)0.14 (+39.68%)0.01 (-67.35%)174.90 (-28.41%)163.42 (-10.36%)167.40 (+9.63%)146.50 (+1.31%)11.19 (-76.17%)
d48746f — 2026-01-12 22:17:100.17 (n/a)0.14 (n/a)0.16 (n/a)0.10 (n/a)0.03 (n/a)244.30 (n/a)182.30 (n/a)152.70 (n/a)144.60 (n/a)46.98 (n/a)

rope_8c_32rows_512cols_32arows_0m

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.82 (-38.66%)0.55 (-39.06%)0.50 (-41.70%)0.42 (-30.52%)0.16 (-41.70%)235.30 (+43.91%)190.06 (+61.92%)198.10 (+71.52%)119.90 (+62.91%)42.58 (+32.73%)
1531781 — 2026-01-23 18:49:241.34 (n/a)0.89 (n/a)0.85 (n/a)0.60 (n/a)0.27 (n/a)163.50 (n/a)117.38 (n/a)115.50 (n/a)73.60 (n/a)32.08 (n/a)

rope_8c_32rows_512cols_8arows_0m

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.38 (-30.44%)0.36 (-29.45%)0.37 (-30.30%)0.34 (-28.67%)0.01 (-48.20%)214.30 (+40.16%)202.92 (+41.59%)200.50 (+43.52%)194.80 (+43.76%)7.75 (+3.85%)
1531781 — 2026-01-23 18:49:240.54 (n/a)0.52 (n/a)0.53 (n/a)0.48 (n/a)0.03 (n/a)152.90 (n/a)143.32 (n/a)139.70 (n/a)135.50 (n/a)7.46 (n/a)

sigmoid_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.08 (-35.88%)0.07 (-32.96%)0.07 (-39.72%)0.05 (-15.10%)0.01 (-58.21%)180.50 (+17.82%)127.22 (+39.93%)113.70 (+65.74%)106.60 (+56.08%)30.29 (-17.70%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.10 (n/a)0.12 (n/a)0.05 (n/a)0.03 (n/a)153.20 (n/a)90.92 (n/a)68.60 (n/a)68.30 (n/a)36.80 (n/a)

sigmoid_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.08 (-32.76%)0.06 (-36.48%)0.06 (-51.73%)0.05 (-21.23%)0.02 (-46.15%)178.80 (+26.99%)136.32 (+51.74%)142.80 (+107.26%)101.30 (+48.75%)32.78 (+2.30%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.10 (n/a)0.12 (n/a)0.06 (n/a)0.03 (n/a)140.80 (n/a)89.84 (n/a)68.90 (n/a)68.10 (n/a)32.04 (n/a)

sigmoid_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.08 (-18.26%)0.07 (-14.51%)0.07 (-27.09%)0.06 (+25.99%)0.01 (-65.32%)134.50 (-20.60%)117.16 (+9.93%)119.90 (+37.19%)102.80 (+22.38%)12.18 (-66.51%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.08 (n/a)0.09 (n/a)0.05 (n/a)0.02 (n/a)169.40 (n/a)106.58 (n/a)87.40 (n/a)84.00 (n/a)36.37 (n/a)

sigmoid_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.07 (-41.40%)0.06 (-40.41%)0.06 (-48.44%)0.04 (-15.52%)0.01 (-65.29%)186.20 (+18.37%)139.14 (+55.05%)133.30 (+94.03%)112.50 (+70.71%)27.98 (-27.98%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.10 (n/a)0.12 (n/a)0.05 (n/a)0.03 (n/a)157.30 (n/a)89.74 (n/a)68.70 (n/a)65.90 (n/a)38.85 (n/a)

silu_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.07 (-31.58%)0.06 (-33.28%)0.05 (-45.76%)0.04 (-25.32%)0.01 (-57.11%)191.90 (+33.91%)150.38 (+43.38%)151.60 (+84.43%)117.80 (+46.15%)27.46 (-14.83%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.08 (n/a)0.10 (n/a)0.06 (n/a)0.02 (n/a)143.30 (n/a)104.88 (n/a)82.20 (n/a)80.60 (n/a)32.24 (n/a)

silu_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.06 (-35.82%)0.05 (-39.23%)0.05 (-50.41%)0.04 (-32.87%)0.01 (-51.10%)206.90 (+48.96%)165.56 (+60.58%)167.80 (+101.68%)127.10 (+55.76%)32.41 (+14.50%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.08 (n/a)0.10 (n/a)0.06 (n/a)0.02 (n/a)138.90 (n/a)103.10 (n/a)83.20 (n/a)81.60 (n/a)28.30 (n/a)

silu_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.08 (-29.65%)0.06 (-41.24%)0.05 (-54.43%)0.04 (-25.85%)0.02 (-32.02%)185.30 (+34.86%)140.20 (+68.19%)153.50 (+119.29%)96.80 (+42.14%)37.56 (+24.08%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.11 (n/a)0.12 (n/a)0.06 (n/a)0.03 (n/a)137.40 (n/a)83.36 (n/a)70.00 (n/a)68.10 (n/a)30.27 (n/a)

silu_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.09 (-22.14%)0.06 (-36.30%)0.05 (-45.28%)0.05 (-37.94%)0.02 (+21.02%)176.90 (+61.11%)146.06 (+63.74%)160.70 (+82.82%)91.30 (+28.41%)36.87 (+154.19%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.09 (n/a)0.09 (n/a)0.07 (n/a)0.02 (n/a)109.80 (n/a)89.20 (n/a)87.90 (n/a)71.10 (n/a)14.50 (n/a)

softmax_1_cols_2_channels_4096_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.21 (+20.19%)0.16 (+13.84%)0.16 (+8.19%)0.12 (+26.61%)0.04 (+17.18%)135.50 (-20.99%)104.68 (-12.65%)102.70 (-7.56%)76.50 (-16.85%)22.55 (-26.21%)
0f0da0b — 2026-01-14 16:31:000.18 (n/a)0.14 (n/a)0.15 (n/a)0.10 (n/a)0.03 (n/a)171.50 (n/a)119.84 (n/a)111.10 (n/a)92.00 (n/a)30.56 (n/a)

softmax_2_cols_2_channels_32768_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:311.09 (n/a)0.92 (n/a)0.83 (n/a)0.79 (n/a)0.15 (n/a)166.40 (n/a)146.10 (n/a)157.40 (n/a)120.40 (n/a)23.07 (n/a)

softmax_2_cols_2_channels_32768_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:311.03 (n/a)0.84 (n/a)0.83 (n/a)0.55 (n/a)0.19 (n/a)237.50 (n/a)164.32 (n/a)158.30 (n/a)127.30 (n/a)44.62 (n/a)

softmax_2_cols_2_channels_4096_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.22 (+40.26%)0.18 (+43.33%)0.20 (+64.08%)0.13 (+24.34%)0.04 (+68.98%)127.80 (-19.62%)94.74 (-28.95%)83.40 (-39.08%)73.90 (-28.67%)24.46 (-5.31%)
0f0da0b — 2026-01-14 16:31:000.16 (n/a)0.13 (n/a)0.12 (n/a)0.10 (n/a)0.03 (n/a)159.00 (n/a)133.34 (n/a)136.90 (n/a)103.60 (n/a)25.83 (n/a)

softmax_2_cols_2_channels_4096_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.24 (+85.21%)0.17 (+69.97%)0.15 (+54.16%)0.11 (+37.21%)0.05 (+203.82%)148.10 (-27.12%)102.78 (-37.88%)106.70 (-35.10%)69.60 (-45.96%)31.81 (+15.98%)
0f0da0b — 2026-01-14 16:31:000.13 (n/a)0.10 (n/a)0.10 (n/a)0.08 (n/a)0.02 (n/a)203.20 (n/a)165.46 (n/a)164.40 (n/a)128.80 (n/a)27.43 (n/a)

softmax_4_cols_4_channels_32768_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:311.05 (n/a)0.75 (n/a)0.80 (n/a)0.45 (n/a)0.24 (n/a)290.20 (n/a)192.58 (n/a)164.70 (n/a)125.40 (n/a)68.37 (n/a)

swiglu

No metrics available.

swiglu_decode_1x2048x2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.00 (-15.46%)0.00 (-27.20%)0.00 (-28.17%)0.00 (-33.90%)0.00 (+1834.83%)5187.20 (+51.28%)4720.84 (+38.38%)4750.90 (+39.22%)4015.10 (+18.29%)390.23 (+3298.45%)
0a6c11c — 2025-12-04 00:39:100.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)3428.80 (n/a)3411.62 (n/a)3412.60 (n/a)3394.20 (n/a)11.48 (n/a)

swiglu_decode_1x2048x2048_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:31n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)
1531781 — 2026-01-23 18:49:240.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)5111.40 (n/a)4324.88 (n/a)4736.10 (n/a)3410.70 (n/a)843.97 (n/a)

swiglu_prefill_256x2048x2048_0

No metrics available.

tanh_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.08 (-14.33%)0.07 (-16.52%)0.06 (-8.26%)0.05 (-16.80%)0.01 (-25.40%)169.30 (+20.24%)129.90 (+19.02%)126.40 (+9.06%)99.90 (+16.71%)26.11 (+11.01%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.08 (n/a)0.07 (n/a)0.06 (n/a)0.02 (n/a)140.80 (n/a)109.14 (n/a)115.90 (n/a)85.60 (n/a)23.52 (n/a)

tanh_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.10 (-2.61%)0.07 (-27.33%)0.07 (-21.16%)0.04 (-46.88%)0.02 (+150.81%)194.50 (+88.29%)133.44 (+51.60%)111.30 (+26.77%)79.90 (+2.70%)48.25 (+400.65%)
1531781 — 2026-01-23 18:49:240.11 (n/a)0.09 (n/a)0.09 (n/a)0.08 (n/a)0.01 (n/a)103.30 (n/a)88.02 (n/a)87.80 (n/a)77.80 (n/a)9.64 (n/a)

tanh_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.07 (-22.83%)0.06 (-33.51%)0.06 (-34.89%)0.04 (-40.11%)0.01 (+7.01%)202.70 (+66.97%)144.36 (+54.56%)132.60 (+53.65%)109.70 (+29.67%)36.72 (+132.32%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.09 (n/a)0.09 (n/a)0.07 (n/a)0.01 (n/a)121.40 (n/a)93.40 (n/a)86.30 (n/a)84.60 (n/a)15.80 (n/a)

tanh_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.07 (-38.66%)0.05 (-46.02%)0.04 (-58.69%)0.04 (-28.22%)0.01 (-51.93%)228.20 (+39.32%)181.62 (+76.50%)194.50 (+141.92%)116.80 (+62.90%)43.72 (+9.35%)
1531781 — 2026-01-23 18:49:240.11 (n/a)0.09 (n/a)0.10 (n/a)0.05 (n/a)0.03 (n/a)163.80 (n/a)102.90 (n/a)80.40 (n/a)71.70 (n/a)39.99 (n/a)

transpose_2048_M_64_N_1_cols_1_channels_64_m_64_n_8_s

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:493.93 (-7.02%)3.68 (+2.48%)3.66 (+1.82%)3.45 (+15.62%)0.15 (-63.64%)151.80 (-13.50%)142.70 (-3.59%)143.40 (-1.78%)133.50 (+7.49%)5.90 (-66.22%)
0a6c11c — 2025-12-04 00:39:104.22 (n/a)3.59 (n/a)3.59 (n/a)2.99 (n/a)0.42 (n/a)175.50 (n/a)148.02 (n/a)146.00 (n/a)124.20 (n/a)17.46 (n/a)

transpose_2048_M_64_N_1_cols_1_channels_64_m_64_n_8_s0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:313.34 (-32.55%)2.93 (-32.53%)2.87 (-34.36%)2.48 (-33.57%)0.32 (-35.36%)211.20 (+50.53%)180.64 (+48.09%)182.60 (+52.29%)157.10 (+48.21%)20.44 (+44.43%)
1531781 — 2026-01-23 18:49:244.95 (n/a)4.34 (n/a)4.37 (n/a)3.74 (n/a)0.50 (n/a)140.30 (n/a)121.98 (n/a)119.90 (n/a)106.00 (n/a)14.15 (n/a)

transpose_2048_M_64_N_1_cols_2_channels_64_m_64_n_8_s

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:495.11 (+15.57%)3.85 (-3.32%)3.40 (-18.64%)3.29 (+7.09%)0.71 (+49.14%)159.60 (-6.61%)140.18 (+4.80%)154.20 (+22.97%)102.60 (-13.49%)22.70 (+19.77%)
0a6c11c — 2025-12-04 00:39:104.42 (n/a)3.99 (n/a)4.18 (n/a)3.07 (n/a)0.48 (n/a)170.90 (n/a)133.76 (n/a)125.40 (n/a)118.60 (n/a)18.95 (n/a)

transpose_2048_M_64_N_1_cols_2_channels_64_m_64_n_8_s0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:314.34 (-28.59%)3.53 (-29.26%)3.77 (-25.99%)2.75 (-32.78%)0.64 (-13.21%)190.50 (+48.71%)152.62 (+42.82%)139.20 (+35.15%)120.90 (+40.09%)28.78 (+83.67%)
1531781 — 2026-01-23 18:49:246.07 (n/a)4.99 (n/a)5.09 (n/a)4.09 (n/a)0.74 (n/a)128.10 (n/a)106.86 (n/a)103.00 (n/a)86.30 (n/a)15.67 (n/a)

weighted_rms_norm_1_cols_2_channels_2048_weights_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.09 (-14.99%)0.08 (-13.55%)0.08 (-17.08%)0.06 (+4.42%)0.01 (-35.00%)222.70 (+43.68%)167.94 (+68.07%)162.10 (+80.92%)134.90 (+76.34%)33.25 (+6.04%)
1531781 — 2026-01-23 18:49:240.11 (n/a)0.09 (n/a)0.09 (n/a)0.05 (n/a)0.02 (n/a)155.00 (n/a)99.92 (n/a)89.60 (n/a)76.50 (n/a)31.36 (n/a)

weighted_rms_norm_2_cols_2_channels_2048_weights_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.09 (-9.05%)0.06 (-26.64%)0.06 (-35.85%)0.05 (+0.20%)0.02 (-23.63%)207.90 (+24.72%)170.10 (+65.60%)172.40 (+94.80%)114.20 (+37.42%)34.68 (-3.30%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.09 (n/a)0.09 (n/a)0.05 (n/a)0.02 (n/a)166.70 (n/a)102.72 (n/a)88.50 (n/a)83.10 (n/a)35.87 (n/a)

weighted_rms_norm_4_cols_2_channels_2048_weights_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.06 (-35.39%)0.05 (-45.77%)0.05 (-49.65%)0.03 (-54.71%)0.01 (+7.03%)295.60 (+148.40%)207.22 (+114.78%)200.70 (+123.25%)155.40 (+74.02%)53.84 (+320.42%)
1531781 — 2026-01-23 18:49:240.09 (n/a)0.09 (n/a)0.09 (n/a)0.07 (n/a)0.01 (n/a)119.00 (n/a)96.48 (n/a)89.90 (n/a)89.30 (n/a)12.81 (n/a)

weighted_rms_norm_8_cols_2_channels_2048_weights_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
02b4f01 — 2026-02-07 00:16:310.06 (-47.83%)0.05 (-52.01%)0.05 (-55.36%)0.04 (-36.10%)0.01 (-61.17%)225.80 (+66.27%)193.78 (+115.50%)190.50 (+138.12%)151.40 (+103.77%)31.13 (+20.83%)
1531781 — 2026-01-23 18:49:240.11 (n/a)0.10 (n/a)0.10 (n/a)0.06 (n/a)0.02 (n/a)135.80 (n/a)89.92 (n/a)80.00 (n/a)74.30 (n/a)25.76 (n/a)

@github-actions
Copy link
Contributor

github-actions bot commented Feb 9, 2026

📊 Test Results for Test Example Applications

4a70412 (2026_02_09_15_08_29)

IRONCLAD

Tested on 2026_02_09_15_08_29 at commit 4a70412.

Test Checks TTFT (mean)TPS (mean)
llama_3.2_1b_prompt_13_tokens_1 ❌ 0/1 n/a n/a
llama_3.2_1b_prompt_13_tokens_40 ❌ 0/1 n/a n/a
llama_3.2_1b_prompt_2048_tokens_1 ❌ 0/1 n/a n/a
llama_3.2_1b_prompt_2048_tokens_40 ❌ 0/1 n/a n/a
📈 Trends (vs main branch) for Test Example Applications

4a70412 (2026_02_09_15_08_29)

IRONCLAD Trends

llama_3.2_1b

Commit/Date Num Tokens (max)Num Tokens (mean)Num Tokens (median)Num Tokens (min)Num Tokens (stddev)TPS (max)TPS (mean)TPS (median)TPS (min)TPS (stddev)TTFT (max)TTFT (mean)TTFT (median)TTFT (min)TTFT (stddev)Total (max)Total (mean)Total (median)Total (min)Total (stddev)
130b6ea — 2025-12-05 21:33:1240.00 (+0.00%)40.00 (+0.00%)40.00 (+0.00%)40.00 (+0.00%)0.00 (n/a)4.71 (-0.42%)4.64 (-0.09%)4.64 (+0.65%)4.55 (-0.22%)0.05 (-17.66%)4.41 (-0.34%)4.39 (-0.19%)4.38 (-0.33%)4.37 (-0.15%)0.01 (-25.90%)12.96 (-0.00%)12.80 (+0.07%)12.80 (-0.23%)12.67 (+0.44%)0.09 (-21.12%)
0a6c11c — 2025-12-03 23:35:1540.00 (n/a)40.00 (n/a)40.00 (n/a)40.00 (n/a)0.00 (n/a)4.73 (n/a)4.64 (n/a)4.61 (n/a)4.56 (n/a)0.06 (n/a)4.42 (n/a)4.40 (n/a)4.40 (n/a)4.37 (n/a)0.02 (n/a)12.96 (n/a)12.79 (n/a)12.83 (n/a)12.62 (n/a)0.12 (n/a)

llama_3.2_1b_prompt_13_tokens_1

Commit/Date Num_Tokens (max)Num_Tokens (mean)Num_Tokens (median)Num_Tokens (min)Num_Tokens (stddev)TPS (max)TPS (mean)TPS (median)TPS (min)TPS (stddev)TTFT (max)TTFT (mean)TTFT (median)TTFT (min)TTFT (stddev)
4a70412 — 2026-02-09 15:08:04n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)
1531781 — 2026-01-23 18:40:341.00 (n/a)1.00 (n/a)1.00 (n/a)1.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.61 (n/a)0.61 (n/a)0.61 (n/a)0.61 (n/a)0.00 (n/a)

llama_3.2_1b_prompt_13_tokens_40

Commit/Date Num_Tokens (max)Num_Tokens (mean)Num_Tokens (median)Num_Tokens (min)Num_Tokens (stddev)TPS (max)TPS (mean)TPS (median)TPS (min)TPS (stddev)TTFT (max)TTFT (mean)TTFT (median)TTFT (min)TTFT (stddev)
4a70412 — 2026-02-09 15:08:04n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)
1531781 — 2026-01-23 18:40:3440.00 (n/a)40.00 (n/a)40.00 (n/a)40.00 (n/a)0.00 (n/a)4.34 (n/a)4.34 (n/a)4.34 (n/a)4.34 (n/a)0.00 (n/a)0.61 (n/a)0.61 (n/a)0.61 (n/a)0.61 (n/a)0.00 (n/a)

llama_3.2_1b_prompt_2048_tokens_1

Commit/Date Num_Tokens (max)Num_Tokens (mean)Num_Tokens (median)Num_Tokens (min)Num_Tokens (stddev)TPS (max)TPS (mean)TPS (median)TPS (min)TPS (stddev)TTFT (max)TTFT (mean)TTFT (median)TTFT (min)TTFT (stddev)
4a70412 — 2026-02-09 15:08:04n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)
1531781 — 2026-01-23 18:40:341.00 (n/a)1.00 (n/a)1.00 (n/a)1.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)2.70 (n/a)2.70 (n/a)2.70 (n/a)2.70 (n/a)0.00 (n/a)

llama_3.2_1b_prompt_2048_tokens_40

Commit/Date Num_Tokens (max)Num_Tokens (mean)Num_Tokens (median)Num_Tokens (min)Num_Tokens (stddev)TPS (max)TPS (mean)TPS (median)TPS (min)TPS (stddev)TTFT (max)TTFT (mean)TTFT (median)TTFT (min)TTFT (stddev)
4a70412 — 2026-02-09 15:08:04n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)
1531781 — 2026-01-23 18:40:3440.00 (n/a)40.00 (n/a)40.00 (n/a)40.00 (n/a)0.00 (n/a)4.22 (n/a)4.22 (n/a)4.22 (n/a)4.22 (n/a)0.00 (n/a)2.70 (n/a)2.70 (n/a)2.70 (n/a)2.70 (n/a)0.00 (n/a)

@github-actions
Copy link
Contributor

github-actions bot commented Feb 9, 2026

📊 Test Results for Small Benchmark/Test Suite

4a70412 (2026_02_09_15_18_52)

IRONCLAD

Tested on 2026_02_09_15_18_52 at commit 4a70412.

Test Checks Latency (mean)Bandwidth (mean)GFLOP/s (mean)
axpy_1_cols_2_channels_2048_tile_2048_3.0_0 ✅ 5/5 124.86 0.10 n/a
axpy_2_cols_2_channels_2048_tile_1024_3.0_0 ✅ 5/5 176.28 0.08 n/a
axpy_4_cols_2_channels_2048_tile_512_3.0_0 ✅ 5/5 142.40 0.09 n/a
axpy_8_cols_2_channels_2048_tile_256_3.0_0 ✅ 5/5 174.96 0.07 n/a
dequant_1_cols_1_channels_2048_tile_2048_0 ✅ 5/5 171.38 0.03 n/a
dequant_1_cols_2_channels_2048_tile_1024_0 ✅ 5/5 136.56 0.04 n/a
dequant_2_cols_1_channels_2048_tile_1024_0 ✅ 5/5 155.56 0.04 n/a
dequant_2_cols_2_channels_2048_tile_512_0 ✅ 5/5 158.54 0.03 n/a
dequant_4_cols_1_channels_2048_tile_512_0 ✅ 5/5 144.88 0.04 n/a
dequant_4_cols_2_channels_2048_tile_256_0 ✅ 5/5 182.16 0.03 n/a
dequant_8_cols_1_channels_2048_tile_256_0 ✅ 5/5 150.28 0.04 n/a
dequant_8_cols_2_channels_2048_tile_128_0 ✅ 5/5 203.08 0.03 n/a
eltwise_add_1_cols_2_channels_2048_tile_2048 ✅ 5/5 150.46 0.09 n/a
eltwise_add_2_cols_2_channels_2048_tile_1024 ✅ 5/5 149.92 0.08 n/a
eltwise_add_4_cols_2_channels_2048_tile_512 ✅ 5/5 166.66 0.08 n/a
eltwise_add_8_cols_2_channels_2048_tile_256 ✅ 5/5 177.82 0.07 n/a
eltwise_mul_1_cols_2_channels_2048_tile_2048 ✅ 5/5 155.98 0.08 n/a
eltwise_mul_2_cols_2_channels_2048_tile_1024 ✅ 5/5 138.18 0.09 n/a
eltwise_mul_4_cols_2_channels_2048_tile_512 ✅ 5/5 186.70 0.07 n/a
eltwise_mul_8_cols_2_channels_2048_tile_256 ✅ 5/5 168.70 0.08 n/a
gelu_1_cols_1_channels_2048_tile_2048 ✅ 5/5 156.12 0.05 n/a
gelu_1_cols_2_channels_2048_tile_1024 ✅ 5/5 179.48 0.05 n/a
gelu_2_cols_1_channels_2048_tile_1024 ✅ 5/5 158.54 0.05 n/a
gelu_2_cols_2_channels_2048_tile_512 ✅ 5/5 196.98 0.04 n/a
gelu_4_cols_1_channels_2048_tile_512 ✅ 5/5 166.02 0.05 n/a
gelu_4_cols_2_channels_2048_tile_256 ✅ 5/5 184.48 0.05 n/a
gelu_8_cols_1_channels_2048_tile_256 ✅ 5/5 186.92 0.04 n/a
gelu_8_cols_2_channels_2048_tile_128 ✅ 5/5 196.22 0.04 n/a
gemm_1792x896x1152_64x32x48_8cols_ccolmaj ✅ 5/5 2328.66 4.08 n/a
gemm_192x384x64_48x96x16_4cols ✅ 5/5 271.90 0.98 n/a
gemm_192x384x64_48x96x16_4cols_bcolmaj_ccolmaj ✅ 5/5 225.16 0.99 n/a
gemm_2048x2048x2048_64x64x64_1cols ✅ 5/5 48621.50 0.52 n/a
gemm_2048x2048x2048_64x64x64_2cols_bcolmaj ✅ 5/5 27699.98 0.91 n/a
gemm_2048x2048x2048_64x64x64_8cols_bcolmaj_ccolmaj ✅ 5/5 7748.14 3.25 n/a
gemm_384x1536x1792_32x48x64_4cols_bcolmaj ✅ 5/5 2026.98 4.00 n/a
gemm_896x1792x640_32x64x80_8cols_ccolmaj ✅ 5/5 1179.90 5.72 n/a
layer_norm_1_cols_1_channels_2048_tile_2048 ✅ 5/5 164.60 0.05 n/a
layer_norm_1_cols_2_channels_2048_tile_1024 ✅ 5/5 142.58 0.06 n/a
layer_norm_2_cols_1_channels_2048_tile_1024 ✅ 5/5 142.80 0.06 n/a
layer_norm_2_cols_2_channels_2048_tile_512 ✅ 5/5 195.94 0.04 n/a
layer_norm_4_cols_1_channels_2048_tile_512 ✅ 5/5 158.08 0.05 n/a
layer_norm_4_cols_2_channels_2048_tile_256 ✅ 5/5 164.32 0.05 n/a
layer_norm_8_cols_1_channels_2048_tile_256 ✅ 5/5 188.84 0.04 n/a
layer_norm_8_cols_2_channels_2048_tile_128 ✅ 5/5 213.76 0.04 n/a
matrix_vector_mul_128x128_32tsi_128tso_1col0 ✅ 5/5 n/a 0.20 n/a
matrix_vector_mul_2048x8192_1tsi_1024tso_2col0 ✅ 5/5 n/a 24.48 n/a
matrix_vector_mul_2048x8192_1tsi_2048tso_1col0 ✅ 5/5 n/a 13.02 n/a
matrix_vector_mul_2048x8192_1tsi_256tso_8col0 ✅ 5/5 n/a 45.39 n/a
matrix_vector_mul_2048x8192_1tsi_512tso_4col0 ✅ 5/5 n/a 40.28 n/a
matrix_vector_mul_8192x2048_4tsi_1024tso_1col0 ✅ 5/5 n/a 13.14 n/a
matrix_vector_mul_8192x2048_4tsi_1024tso_2col0 ✅ 5/5 n/a 24.34 n/a
matrix_vector_mul_8192x2048_4tsi_1024tso_4col0 ✅ 5/5 n/a 39.85 n/a
matrix_vector_mul_8192x2048_4tsi_1024tso_8col0 ✅ 5/5 n/a 43.26 n/a
mem_copy_16_cores_2_chans_2048_tile_128_False0 ✅ 5/5 166.04 0.05 n/a
mem_copy_1_cores_1_chans_2048_tile_2048_False0 ✅ 5/5 125.66 0.07 n/a
mem_copy_2_cores_1_chans_2048_tile_1024_False0 ✅ 5/5 127.56 0.06 n/a
mem_copy_2_cores_2_chans_2048_tile_1024_False0 ✅ 5/5 128.28 0.06 n/a
mem_copy_4_cores_1_chans_2048_tile_512_False0 ✅ 5/5 133.14 0.06 n/a
mem_copy_4_cores_2_chans_2048_tile_512_False0 ✅ 5/5 139.38 0.06 n/a
mem_copy_8_cores_1_chans_2048_tile_256_False0 ✅ 5/5 131.68 0.06 n/a
mem_copy_8_cores_2_chans_2048_tile_256_False0 ✅ 5/5 170.04 0.05 n/a
mha0 ✅ 5/5 47452.78 0.18 n/a
relu_1_cols_1_channels_2048_tile_2048 ✅ 5/5 149.80 0.06 n/a
relu_2_cols_1_channels_2048_tile_1024 ✅ 5/5 138.20 0.06 n/a
relu_4_cols_1_channels_2048_tile_512 ✅ 5/5 173.56 0.05 n/a
relu_8_cols_1_channels_2048_tile_256 ✅ 5/5 192.02 0.05 n/a
rms_norm_1_cols_1_channels_2048_tile_2048 ✅ 5/5 160.50 0.05 n/a
rms_norm_1_cols_2_channels_2048_tile_1024 ✅ 5/5 220.28 0.04 n/a
rms_norm_2_cols_1_channels_2048_tile_1024 ✅ 5/5 176.82 0.05 n/a
rms_norm_2_cols_2_channels_2048_tile_512 ✅ 5/5 182.54 0.05 n/a
rms_norm_4_cols_1_channels_2048_tile_512 ✅ 5/5 181.68 0.05 n/a
rms_norm_4_cols_2_channels_2048_tile_256 ✅ 5/5 202.66 0.04 n/a
rms_norm_8_cols_1_channels_2048_tile_256 ✅ 5/5 191.42 0.04 n/a
rms_norm_8_cols_2_channels_2048_tile_128 ✅ 5/5 232.08 0.04 n/a
rope_1c_32rows_512cols_32arows_0m ✅ 5/5 167.64 0.61 n/a
rope_1c_32rows_512cols_8arows_0m ✅ 5/5 178.26 0.44 n/a
rope_2c_32rows_512cols_32arows_0m ✅ 5/5 145.34 0.69 n/a
rope_2c_32rows_512cols_8arows_0m ✅ 5/5 186.34 0.42 n/a
rope_8c_32rows_512cols_32arows_0m ✅ 5/5 212.48 0.47 n/a
rope_8c_32rows_512cols_8arows_0m ✅ 5/5 167.26 0.47 n/a
sigmoid_1_cols_1_channels_2048_tile_2048 ✅ 5/5 150.16 0.05 n/a
sigmoid_2_cols_1_channels_2048_tile_1024 ✅ 5/5 145.46 0.06 n/a
sigmoid_4_cols_1_channels_2048_tile_512 ✅ 5/5 140.96 0.06 n/a
sigmoid_8_cols_1_channels_2048_tile_256 ✅ 5/5 161.88 0.05 n/a
silu_1_cols_1_channels_2048_tile_2048 ✅ 5/5 152.52 0.06 n/a
silu_2_cols_1_channels_2048_tile_1024 ✅ 5/5 167.56 0.05 n/a
silu_4_cols_1_channels_2048_tile_512 ✅ 5/5 171.18 0.05 n/a
silu_8_cols_1_channels_2048_tile_256 ✅ 5/5 162.00 0.05 n/a
softmax_2_cols_2_channels_32768_tile_1024 ✅ 5/5 143.86 0.93 n/a
softmax_2_cols_2_channels_32768_tile_512 ✅ 5/5 158.20 0.84 n/a
softmax_4_cols_4_channels_32768_tile_2048 ✅ 5/5 168.34 0.79 n/a
swiglu_decode_1x2048x2048_0 ✅ 5/5 n/a n/a n/a
swiglu_prefill_256x2048x2048_0 ❌ 0/5 n/a n/a n/a
tanh_1_cols_1_channels_2048_tile_2048 ✅ 5/5 154.24 0.06 n/a
tanh_2_cols_1_channels_2048_tile_1024 ✅ 5/5 174.26 0.05 n/a
tanh_4_cols_1_channels_2048_tile_512 ✅ 5/5 148.96 0.06 n/a
tanh_8_cols_1_channels_2048_tile_256 ✅ 5/5 181.22 0.05 n/a
transpose_2048_M_64_N_1_cols_1_channels_64_m_64_n_8_s0 ✅ 5/5 169.48 3.16 n/a
transpose_2048_M_64_N_1_cols_2_channels_64_m_64_n_8_s0 ✅ 5/5 175.16 3.05 n/a
weighted_rms_norm_1_cols_2_channels_2048_weights_2048 ✅ 5/5 161.74 0.08 n/a
weighted_rms_norm_2_cols_2_channels_2048_weights_1024 ✅ 5/5 172.66 0.06 n/a
weighted_rms_norm_4_cols_2_channels_2048_weights_512 ✅ 5/5 183.56 0.05 n/a
weighted_rms_norm_8_cols_2_channels_2048_weights_256 ✅ 5/5 202.26 0.04 n/a
📈 Trends (vs main branch) for Small Benchmark/Test Suite

4a70412 (2026_02_09_15_18_52)

IRONCLAD Trends

axpy_1_cols_2_channels_2048_tile_2048_3.0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.13 (-1.96%)0.11 (+2.52%)0.10 (+1.86%)0.08 (-1.91%)0.02 (-1.72%)162.40 (+1.95%)120.96 (-2.36%)117.00 (-1.85%)94.40 (+2.05%)23.11 (+4.09%)
0a6c11c — 2025-12-04 00:39:100.13 (n/a)0.10 (n/a)0.10 (n/a)0.08 (n/a)0.02 (n/a)159.30 (n/a)123.88 (n/a)119.20 (n/a)92.50 (n/a)22.21 (n/a)

axpy_1_cols_2_channels_2048_tile_2048_3.0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.11 (-29.34%)0.10 (-21.85%)0.10 (-10.40%)0.09 (-16.69%)0.01 (-63.47%)142.60 (+20.03%)124.86 (+24.49%)123.30 (+11.68%)110.40 (+41.54%)13.00 (-36.58%)
1531781 — 2026-01-23 18:49:240.16 (n/a)0.13 (n/a)0.11 (n/a)0.10 (n/a)0.03 (n/a)118.80 (n/a)100.30 (n/a)110.40 (n/a)78.00 (n/a)20.49 (n/a)

axpy_2_cols_2_channels_2048_tile_1024_3.0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.11 (-26.77%)0.09 (-9.09%)0.09 (-10.81%)0.08 (+8.25%)0.01 (-55.42%)152.00 (-7.60%)135.46 (+6.28%)142.80 (+12.18%)115.00 (+36.58%)15.03 (-41.36%)
0a6c11c — 2025-12-04 00:39:100.15 (n/a)0.10 (n/a)0.10 (n/a)0.07 (n/a)0.02 (n/a)164.50 (n/a)127.46 (n/a)127.30 (n/a)84.20 (n/a)25.64 (n/a)

axpy_2_cols_2_channels_2048_tile_1024_3.0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.10 (-31.56%)0.08 (-38.73%)0.08 (-44.81%)0.04 (-44.37%)0.02 (-16.96%)282.00 (+79.73%)176.28 (+68.75%)163.60 (+81.17%)128.50 (+46.19%)62.74 (+112.55%)
1531781 — 2026-01-23 18:49:240.14 (n/a)0.12 (n/a)0.14 (n/a)0.08 (n/a)0.03 (n/a)156.90 (n/a)104.46 (n/a)90.30 (n/a)87.90 (n/a)29.52 (n/a)

axpy_4_cols_2_channels_2048_tile_512_3.0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.12 (-10.21%)0.10 (-8.41%)0.10 (-6.20%)0.07 (+3.52%)0.02 (-23.48%)166.70 (-3.36%)128.78 (+7.33%)121.80 (+6.65%)103.90 (+11.36%)22.95 (-18.82%)
0a6c11c — 2025-12-04 00:39:100.13 (n/a)0.11 (n/a)0.11 (n/a)0.07 (n/a)0.02 (n/a)172.50 (n/a)119.98 (n/a)114.20 (n/a)93.30 (n/a)28.27 (n/a)

axpy_4_cols_2_channels_2048_tile_512_3.0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.14 (-12.99%)0.09 (-34.02%)0.08 (-42.84%)0.07 (-25.71%)0.03 (+0.52%)180.50 (+34.60%)142.40 (+54.05%)145.90 (+74.94%)90.40 (+15.01%)33.74 (+44.31%)
1531781 — 2026-01-23 18:49:240.16 (n/a)0.14 (n/a)0.15 (n/a)0.09 (n/a)0.03 (n/a)134.10 (n/a)92.44 (n/a)83.40 (n/a)78.60 (n/a)23.38 (n/a)

axpy_8_cols_2_channels_2048_tile_256_3.0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.10 (-16.19%)0.08 (-10.40%)0.09 (-2.09%)0.07 (-17.38%)0.01 (-10.61%)184.40 (+21.00%)153.18 (+11.99%)144.10 (+2.20%)127.10 (+19.34%)21.26 (+34.76%)
0a6c11c — 2025-12-04 00:39:100.12 (n/a)0.09 (n/a)0.09 (n/a)0.08 (n/a)0.01 (n/a)152.40 (n/a)136.78 (n/a)141.00 (n/a)106.50 (n/a)15.78 (n/a)

axpy_8_cols_2_channels_2048_tile_256_3.0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.09 (-47.86%)0.07 (-49.79%)0.07 (-53.55%)0.06 (-30.41%)0.01 (-68.12%)211.00 (+43.73%)174.96 (+89.68%)172.10 (+115.12%)138.80 (+91.98%)26.18 (-15.72%)
1531781 — 2026-01-23 18:49:240.17 (n/a)0.14 (n/a)0.15 (n/a)0.08 (n/a)0.03 (n/a)146.80 (n/a)92.24 (n/a)80.00 (n/a)72.30 (n/a)31.07 (n/a)

dequant_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.05 (-18.83%)0.04 (-5.21%)0.04 (-2.69%)0.03 (+37.10%)0.00 (-53.98%)160.40 (-27.06%)141.12 (-0.91%)136.40 (+2.79%)114.30 (+23.17%)17.37 (-59.06%)
0a6c11c — 2025-12-04 00:39:100.06 (n/a)0.04 (n/a)0.04 (n/a)0.02 (n/a)0.01 (n/a)219.90 (n/a)142.42 (n/a)132.70 (n/a)92.80 (n/a)42.43 (n/a)

dequant_1_cols_1_channels_2048_tile_2048_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.04 (-31.68%)0.03 (-32.38%)0.03 (-33.50%)0.03 (-37.64%)0.01 (-22.10%)201.20 (+60.32%)171.38 (+48.92%)182.90 (+50.41%)129.00 (+46.42%)28.19 (+83.02%)
1531781 — 2026-01-23 18:49:240.06 (n/a)0.05 (n/a)0.04 (n/a)0.04 (n/a)0.01 (n/a)125.50 (n/a)115.08 (n/a)121.60 (n/a)88.10 (n/a)15.40 (n/a)

dequant_1_cols_2_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.06 (-13.28%)0.05 (-8.93%)0.06 (-13.59%)0.04 (+4.63%)0.01 (-39.10%)133.60 (-4.43%)106.08 (+5.85%)95.10 (+15.69%)87.90 (+15.35%)17.90 (-31.97%)
0a6c11c — 2025-12-04 00:39:100.07 (n/a)0.06 (n/a)0.06 (n/a)0.04 (n/a)0.01 (n/a)139.80 (n/a)100.22 (n/a)82.20 (n/a)76.20 (n/a)26.31 (n/a)

dequant_1_cols_2_channels_2048_tile_1024_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.05 (-31.43%)0.04 (-19.41%)0.04 (-19.36%)0.03 (-0.18%)0.01 (-55.11%)164.00 (+0.18%)136.56 (+19.10%)138.00 (+23.99%)111.60 (+45.88%)20.54 (-35.04%)
1531781 — 2026-01-23 18:49:240.07 (n/a)0.05 (n/a)0.05 (n/a)0.03 (n/a)0.01 (n/a)163.70 (n/a)114.66 (n/a)111.30 (n/a)76.50 (n/a)31.63 (n/a)

dequant_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.07 (+5.66%)0.05 (+1.21%)0.05 (+0.61%)0.04 (-15.02%)0.01 (+58.42%)143.40 (+17.64%)101.02 (+1.49%)97.60 (-0.61%)78.90 (-5.40%)22.56 (+78.52%)
0a6c11c — 2025-12-04 00:39:100.06 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)121.90 (n/a)99.54 (n/a)98.20 (n/a)83.40 (n/a)12.64 (n/a)

dequant_2_cols_1_channels_2048_tile_1024_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.04 (-13.34%)0.04 (-25.79%)0.04 (-25.69%)0.02 (-30.60%)0.01 (+10.94%)212.90 (+44.05%)155.56 (+37.49%)143.30 (+34.55%)117.10 (+15.48%)36.37 (+85.41%)
1531781 — 2026-01-23 18:49:240.05 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)147.80 (n/a)113.14 (n/a)106.50 (n/a)101.40 (n/a)19.61 (n/a)

dequant_2_cols_2_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.05 (+17.06%)0.04 (+8.98%)0.05 (+21.37%)0.03 (-11.65%)0.01 (+301.89%)159.70 (+13.18%)126.32 (-5.68%)109.00 (-17.67%)106.80 (-14.56%)22.48 (+280.15%)
0a6c11c — 2025-12-04 00:39:100.04 (n/a)0.04 (n/a)0.04 (n/a)0.04 (n/a)0.00 (n/a)141.10 (n/a)133.92 (n/a)132.40 (n/a)125.00 (n/a)5.91 (n/a)

dequant_2_cols_2_channels_2048_tile_512_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.05 (-11.99%)0.03 (-22.01%)0.03 (-31.37%)0.02 (-24.95%)0.01 (+1.86%)211.00 (+33.29%)158.54 (+30.61%)154.70 (+45.67%)111.60 (+13.65%)40.36 (+54.50%)
1531781 — 2026-01-23 18:49:240.05 (n/a)0.04 (n/a)0.05 (n/a)0.03 (n/a)0.01 (n/a)158.30 (n/a)121.38 (n/a)106.20 (n/a)98.20 (n/a)26.13 (n/a)

dequant_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.06 (+20.24%)0.05 (-3.19%)0.05 (-6.73%)0.03 (-20.69%)0.01 (+194.26%)156.10 (+26.09%)120.86 (+7.15%)114.90 (+7.18%)88.30 (-16.85%)24.40 (+214.82%)
0a6c11c — 2025-12-04 00:39:100.05 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.00 (n/a)123.80 (n/a)112.80 (n/a)107.20 (n/a)106.20 (n/a)7.75 (n/a)

dequant_4_cols_1_channels_2048_tile_512_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.05 (-20.23%)0.04 (-21.64%)0.04 (-19.81%)0.03 (-14.89%)0.01 (-19.35%)193.60 (+17.48%)144.88 (+27.51%)128.70 (+24.71%)106.30 (+25.35%)39.46 (+21.22%)
1531781 — 2026-01-23 18:49:240.06 (n/a)0.05 (n/a)0.05 (n/a)0.03 (n/a)0.01 (n/a)164.80 (n/a)113.62 (n/a)103.20 (n/a)84.80 (n/a)32.55 (n/a)

dequant_4_cols_2_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.05 (-6.09%)0.04 (-8.04%)0.04 (-12.46%)0.03 (-0.41%)0.01 (-20.19%)168.10 (+0.42%)135.74 (+7.44%)132.20 (+14.26%)107.40 (+6.44%)22.91 (-12.67%)
0a6c11c — 2025-12-04 00:39:100.05 (n/a)0.04 (n/a)0.05 (n/a)0.03 (n/a)0.01 (n/a)167.40 (n/a)126.34 (n/a)115.70 (n/a)100.90 (n/a)26.23 (n/a)

dequant_4_cols_2_channels_2048_tile_256_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.04 (-34.49%)0.03 (-40.72%)0.03 (-47.52%)0.02 (-17.00%)0.01 (-45.49%)227.00 (+20.49%)182.16 (+61.60%)178.80 (+90.62%)121.60 (+52.57%)40.20 (-7.76%)
1531781 — 2026-01-23 18:49:240.07 (n/a)0.05 (n/a)0.06 (n/a)0.03 (n/a)0.01 (n/a)188.40 (n/a)112.72 (n/a)93.80 (n/a)79.70 (n/a)43.58 (n/a)

dequant_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.04 (-13.65%)0.04 (-13.67%)0.04 (-18.56%)0.03 (-13.52%)0.00 (-23.88%)170.40 (+15.68%)140.12 (+15.33%)135.80 (+22.78%)119.60 (+15.78%)17.94 (+1.87%)
0a6c11c — 2025-12-04 00:39:100.05 (n/a)0.04 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)147.30 (n/a)121.50 (n/a)110.60 (n/a)103.30 (n/a)17.61 (n/a)

dequant_8_cols_1_channels_2048_tile_256_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.04 (-4.79%)0.04 (-7.43%)0.03 (-12.18%)0.03 (-1.79%)0.01 (+8.51%)170.00 (+1.86%)150.28 (+8.38%)154.70 (+13.83%)123.50 (+5.02%)20.82 (+16.05%)
1531781 — 2026-01-23 18:49:240.04 (n/a)0.04 (n/a)0.04 (n/a)0.03 (n/a)0.00 (n/a)166.90 (n/a)138.66 (n/a)135.90 (n/a)117.60 (n/a)17.94 (n/a)

dequant_8_cols_2_channels_2048_tile_128

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.05 (+17.21%)0.03 (-1.70%)0.03 (-3.04%)0.03 (-16.20%)0.01 (+95.59%)198.20 (+19.33%)160.00 (+4.92%)166.20 (+3.17%)106.50 (-14.66%)29.86 (+87.19%)
0a6c11c — 2025-12-04 00:39:100.04 (n/a)0.03 (n/a)0.03 (n/a)0.03 (n/a)0.00 (n/a)166.10 (n/a)152.50 (n/a)161.10 (n/a)124.80 (n/a)15.95 (n/a)

dequant_8_cols_2_channels_2048_tile_128_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.03 (-41.26%)0.03 (-32.17%)0.03 (-33.41%)0.02 (-9.72%)0.00 (-66.18%)254.60 (+10.74%)203.08 (+39.48%)196.70 (+50.15%)180.70 (+70.15%)30.12 (-38.58%)
1531781 — 2026-01-23 18:49:240.05 (n/a)0.04 (n/a)0.04 (n/a)0.02 (n/a)0.01 (n/a)229.90 (n/a)145.60 (n/a)131.00 (n/a)106.20 (n/a)49.03 (n/a)

eltwise_add_1_cols_2_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.10 (-39.07%)0.09 (-29.24%)0.09 (-10.88%)0.05 (-31.13%)0.02 (-51.52%)226.30 (+45.25%)150.46 (+37.06%)135.20 (+12.20%)121.10 (+64.09%)43.42 (+26.28%)
1531781 — 2026-01-23 18:49:240.17 (n/a)0.12 (n/a)0.10 (n/a)0.08 (n/a)0.04 (n/a)155.80 (n/a)109.78 (n/a)120.50 (n/a)73.80 (n/a)34.38 (n/a)

eltwise_add_2_cols_2_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.11 (-29.05%)0.08 (-33.54%)0.08 (-38.08%)0.07 (-19.15%)0.02 (-38.61%)181.40 (+23.65%)149.92 (+48.14%)148.10 (+61.50%)114.70 (+40.91%)26.53 (+2.05%)
1531781 — 2026-01-23 18:49:240.15 (n/a)0.13 (n/a)0.13 (n/a)0.08 (n/a)0.03 (n/a)146.70 (n/a)101.20 (n/a)91.70 (n/a)81.40 (n/a)26.00 (n/a)

eltwise_add_4_cols_2_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.09 (-47.59%)0.08 (-48.56%)0.07 (-53.26%)0.06 (-38.86%)0.01 (-52.98%)189.80 (+63.62%)166.66 (+92.76%)173.10 (+113.97%)138.80 (+90.92%)24.56 (+42.60%)
1531781 — 2026-01-23 18:49:240.17 (n/a)0.15 (n/a)0.15 (n/a)0.11 (n/a)0.02 (n/a)116.00 (n/a)86.46 (n/a)80.90 (n/a)72.70 (n/a)17.22 (n/a)

eltwise_add_8_cols_2_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.09 (-47.62%)0.07 (-49.48%)0.07 (-50.88%)0.06 (-47.82%)0.01 (-50.00%)209.80 (+91.60%)177.82 (+97.58%)175.00 (+103.49%)143.80 (+90.97%)24.15 (+79.72%)
1531781 — 2026-01-23 18:49:240.16 (n/a)0.14 (n/a)0.14 (n/a)0.11 (n/a)0.02 (n/a)109.50 (n/a)90.00 (n/a)86.00 (n/a)75.30 (n/a)13.44 (n/a)

eltwise_mul_1_cols_2_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.09 (-32.40%)0.08 (-36.87%)0.08 (-38.75%)0.06 (-38.46%)0.01 (-17.66%)191.90 (+62.49%)155.98 (+59.36%)151.20 (+63.28%)133.80 (+48.01%)22.75 (+96.97%)
1531781 — 2026-01-23 18:49:240.14 (n/a)0.13 (n/a)0.13 (n/a)0.10 (n/a)0.01 (n/a)118.10 (n/a)97.88 (n/a)92.60 (n/a)90.40 (n/a)11.55 (n/a)

eltwise_mul_2_cols_2_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.10 (-22.81%)0.09 (-26.42%)0.09 (-34.53%)0.08 (-17.60%)0.01 (-30.06%)157.80 (+21.38%)138.18 (+35.34%)142.60 (+52.84%)118.00 (+29.53%)17.96 (+8.66%)
1531781 — 2026-01-23 18:49:240.13 (n/a)0.12 (n/a)0.13 (n/a)0.09 (n/a)0.02 (n/a)130.00 (n/a)102.10 (n/a)93.30 (n/a)91.10 (n/a)16.53 (n/a)

eltwise_mul_4_cols_2_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.09 (-47.44%)0.07 (-53.94%)0.07 (-56.70%)0.06 (-45.71%)0.01 (-47.54%)210.00 (+84.21%)186.70 (+116.64%)188.10 (+130.80%)141.10 (+90.16%)27.76 (+75.13%)
1531781 — 2026-01-23 18:49:240.17 (n/a)0.15 (n/a)0.15 (n/a)0.11 (n/a)0.02 (n/a)114.00 (n/a)86.18 (n/a)81.50 (n/a)74.20 (n/a)15.85 (n/a)

eltwise_mul_8_cols_2_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.09 (-26.50%)0.08 (-33.20%)0.07 (-37.34%)0.05 (-31.28%)0.01 (-23.04%)225.30 (+45.54%)168.70 (+50.25%)166.90 (+59.56%)129.90 (+36.02%)35.96 (+49.00%)
1531781 — 2026-01-23 18:49:240.13 (n/a)0.11 (n/a)0.12 (n/a)0.08 (n/a)0.02 (n/a)154.80 (n/a)112.28 (n/a)104.60 (n/a)95.50 (n/a)24.14 (n/a)

gelu_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.06 (-43.38%)0.05 (-45.82%)0.05 (-51.77%)0.04 (-53.36%)0.01 (-40.89%)217.90 (+114.47%)156.12 (+86.30%)150.20 (+107.46%)127.60 (+76.73%)36.21 (+129.88%)
1531781 — 2026-01-23 18:49:240.11 (n/a)0.10 (n/a)0.11 (n/a)0.08 (n/a)0.02 (n/a)101.60 (n/a)83.80 (n/a)72.40 (n/a)72.20 (n/a)15.75 (n/a)

gelu_1_cols_2_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.06 (-50.26%)0.05 (-55.85%)0.04 (-63.52%)0.04 (-47.65%)0.01 (-58.46%)210.00 (+90.91%)179.48 (+123.62%)189.30 (+173.95%)133.50 (+101.05%)28.96 (+56.22%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.11 (n/a)0.12 (n/a)0.07 (n/a)0.02 (n/a)110.00 (n/a)80.26 (n/a)69.10 (n/a)66.40 (n/a)18.54 (n/a)

gelu_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.06 (-54.16%)0.05 (-50.10%)0.05 (-56.97%)0.05 (-18.27%)0.00 (-87.80%)168.60 (+22.35%)158.54 (+86.47%)158.50 (+132.40%)147.80 (+117.99%)9.92 (-67.20%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.10 (n/a)0.12 (n/a)0.06 (n/a)0.03 (n/a)137.80 (n/a)85.02 (n/a)68.20 (n/a)67.80 (n/a)30.24 (n/a)

gelu_2_cols_2_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.06 (-51.42%)0.04 (-59.46%)0.04 (-65.68%)0.03 (-57.49%)0.01 (-30.49%)253.70 (+135.34%)196.98 (+156.48%)202.70 (+191.65%)138.00 (+105.66%)56.56 (+225.07%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.11 (n/a)0.12 (n/a)0.08 (n/a)0.02 (n/a)107.80 (n/a)76.80 (n/a)69.50 (n/a)67.10 (n/a)17.40 (n/a)

gelu_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.06 (-55.26%)0.05 (-52.27%)0.05 (-54.52%)0.04 (-32.78%)0.01 (-71.56%)227.70 (+48.73%)166.02 (+94.04%)152.30 (+119.77%)140.90 (+123.65%)35.08 (-7.53%)
1531781 — 2026-01-23 18:49:240.13 (n/a)0.11 (n/a)0.12 (n/a)0.05 (n/a)0.03 (n/a)153.10 (n/a)85.56 (n/a)69.30 (n/a)63.00 (n/a)37.94 (n/a)

gelu_4_cols_2_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.06 (-45.20%)0.05 (-56.39%)0.04 (-63.38%)0.04 (-50.61%)0.01 (-40.45%)223.60 (+102.35%)184.48 (+131.18%)192.40 (+172.91%)127.80 (+82.57%)37.72 (+115.77%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.11 (n/a)0.12 (n/a)0.07 (n/a)0.02 (n/a)110.50 (n/a)79.80 (n/a)70.50 (n/a)70.00 (n/a)17.48 (n/a)

gelu_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.05 (-58.22%)0.04 (-55.54%)0.05 (-58.48%)0.04 (-2.73%)0.01 (-84.74%)214.10 (+2.83%)186.92 (+89.31%)173.10 (+140.75%)168.10 (+139.46%)22.67 (-62.96%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.10 (n/a)0.11 (n/a)0.04 (n/a)0.03 (n/a)208.20 (n/a)98.74 (n/a)71.90 (n/a)70.20 (n/a)61.20 (n/a)

gelu_8_cols_2_channels_2048_tile_128

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.05 (-54.63%)0.04 (-53.75%)0.04 (-55.68%)0.04 (-40.08%)0.00 (-77.54%)222.70 (+66.94%)196.22 (+110.00%)185.70 (+125.64%)180.50 (+120.39%)18.35 (-18.32%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.09 (n/a)0.10 (n/a)0.06 (n/a)0.02 (n/a)133.40 (n/a)93.44 (n/a)82.30 (n/a)81.90 (n/a)22.47 (n/a)

gemm_1792x896x1152_64x32x48_8cols_ccolmaj

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
4a70412 — 2026-02-09 15:12:374.94 (-4.13%)4.08 (-16.58%)3.90 (-21.89%)3.63 (-15.05%)0.51 (+42.38%)2591.80 (+17.72%)2328.66 (+20.66%)2413.30 (+28.03%)1905.40 (+4.31%)260.49 (+69.10%)1941.48 (-4.13%)1606.38 (-16.58%)1532.93 (-21.89%)1427.31 (-15.05%)199.29 (+42.38%)
1531781 — 2026-01-23 18:49:245.15 (n/a)4.90 (n/a)4.99 (n/a)4.27 (n/a)0.36 (n/a)2201.70 (n/a)1929.94 (n/a)1885.00 (n/a)1826.70 (n/a)154.05 (n/a)2025.19 (n/a)1925.76 (n/a)1962.51 (n/a)1680.20 (n/a)139.97 (n/a)

gemm_192x384x64_48x96x16_4cols

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
4a70412 — 2026-02-09 15:12:371.76 (-25.25%)0.98 (-47.25%)0.99 (-44.57%)0.55 (-58.01%)0.49 (+25.51%)399.60 (+138.14%)271.90 (+118.85%)224.40 (+80.39%)125.40 (+33.83%)119.55 (+325.26%)75.25 (-25.25%)41.64 (-47.25%)42.06 (-44.57%)23.62 (-58.01%)21.00 (+25.51%)
1531781 — 2026-01-23 18:49:242.36 (n/a)1.85 (n/a)1.78 (n/a)1.32 (n/a)0.39 (n/a)167.80 (n/a)124.24 (n/a)124.40 (n/a)93.70 (n/a)28.11 (n/a)100.67 (n/a)78.94 (n/a)75.88 (n/a)56.25 (n/a)16.73 (n/a)

gemm_192x384x64_48x96x16_4cols_bcolmaj_ccolmaj

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
4a70412 — 2026-02-09 15:12:371.05 (-52.69%)0.99 (-52.09%)1.01 (-54.19%)0.90 (-43.57%)0.06 (-76.33%)244.90 (+77.21%)225.16 (+106.19%)219.30 (+118.21%)211.00 (+111.42%)14.61 (-11.68%)44.73 (-52.69%)42.05 (-52.09%)43.02 (-54.19%)38.54 (-43.57%)2.67 (-76.33%)
1531781 — 2026-01-23 18:49:242.22 (n/a)2.06 (n/a)2.20 (n/a)1.60 (n/a)0.26 (n/a)138.20 (n/a)109.20 (n/a)100.50 (n/a)99.80 (n/a)16.54 (n/a)94.54 (n/a)87.77 (n/a)93.91 (n/a)68.30 (n/a)11.26 (n/a)

gemm_2048x2048x2048_64x64x32_8_cols_0_bcolmaj_0_ccolmaj_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)GFLOP/s (max)GFLOP/s (mean)GFLOP/s (median)GFLOP/s (min)GFLOP/s (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3a053d3 — 2025-11-26 01:12:163.72 (-0.11%)3.55 (-2.75%)3.56 (-2.75%)3.36 (-5.33%)0.12 (+76.10%)2725.23 (-0.11%)2602.52 (-2.75%)2612.11 (-2.75%)2459.89 (-5.33%)86.27 (+76.10%)6984.00 (+5.63%)6608.60 (+2.91%)6577.00 (+2.83%)6304.00 (+0.11%)221.87 (+87.15%)
9e98049 — 2025-11-25 20:50:553.72 (n/a)3.65 (n/a)3.66 (n/a)3.54 (n/a)0.07 (n/a)2728.26 (n/a)2676.06 (n/a)2686.03 (n/a)2598.29 (n/a)48.99 (n/a)6612.00 (n/a)6422.00 (n/a)6396.00 (n/a)6297.00 (n/a)118.55 (n/a)

gemm_2048x2048x2048_64x64x32_8_cols_0_bcolmaj_1_ccolmaj_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)GFLOP/s (max)GFLOP/s (mean)GFLOP/s (median)GFLOP/s (min)GFLOP/s (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3a053d3 — 2025-11-26 01:12:163.40 (-3.14%)3.38 (-0.82%)3.38 (-0.82%)3.35 (+4.93%)0.02 (-84.51%)2493.81 (-3.14%)2475.27 (-0.82%)2479.06 (-0.82%)2453.92 (+4.93%)13.33 (-84.51%)7001.00 (-4.70%)6940.80 (+0.71%)6930.00 (+0.83%)6889.00 (+3.24%)37.43 (-84.76%)
9e98049 — 2025-11-25 20:50:553.51 (n/a)3.40 (n/a)3.41 (n/a)3.19 (n/a)0.12 (n/a)2574.53 (n/a)2495.72 (n/a)2499.62 (n/a)2338.67 (n/a)86.02 (n/a)7346.00 (n/a)6892.20 (n/a)6873.00 (n/a)6673.00 (n/a)245.57 (n/a)

gemm_2048x2048x2048_64x64x32_8_cols_1_bcolmaj_0_ccolmaj_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)GFLOP/s (max)GFLOP/s (mean)GFLOP/s (median)GFLOP/s (min)GFLOP/s (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3a053d3 — 2025-11-26 01:12:163.02 (+0.52%)2.87 (-2.05%)2.86 (-2.24%)2.70 (-6.16%)0.11 (+131.66%)2216.19 (+0.52%)2104.44 (-2.05%)2094.08 (-2.24%)1980.84 (-6.16%)80.25 (+131.67%)8673.00 (+6.56%)8175.60 (+2.21%)8204.00 (+2.29%)7752.00 (-0.51%)314.03 (+145.36%)
9e98049 — 2025-11-25 20:50:553.01 (n/a)2.93 (n/a)2.92 (n/a)2.88 (n/a)0.05 (n/a)2204.81 (n/a)2148.41 (n/a)2142.13 (n/a)2110.81 (n/a)34.64 (n/a)8139.00 (n/a)7998.60 (n/a)8020.00 (n/a)7792.00 (n/a)127.99 (n/a)

gemm_2048x2048x2048_64x64x64_1cols

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
4a70412 — 2026-02-09 15:12:370.52 (-0.43%)0.52 (-0.52%)0.52 (-0.45%)0.52 (-0.73%)0.00 (+199.48%)48769.50 (+0.73%)48621.50 (+0.52%)48578.70 (+0.45%)48554.20 (+0.43%)89.04 (+202.91%)353.83 (-0.43%)353.34 (-0.52%)353.65 (-0.45%)352.27 (-0.73%)0.65 (+199.46%)
1531781 — 2026-01-23 18:49:240.52 (n/a)0.52 (n/a)0.52 (n/a)0.52 (n/a)0.00 (n/a)48415.10 (n/a)48370.52 (n/a)48362.20 (n/a)48344.40 (n/a)29.40 (n/a)355.36 (n/a)355.17 (n/a)355.23 (n/a)354.85 (n/a)0.22 (n/a)

gemm_2048x2048x2048_64x64x64_2_cols_0_bcolmaj_0_ccolmaj_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:491.04 (+0.94%)1.03 (+0.23%)1.03 (+0.09%)1.02 (+0.07%)0.00 (+197.51%)24576.00 (-0.07%)24485.82 (-0.23%)24530.40 (-0.09%)24266.40 (-0.94%)111.26 (+194.35%)
0a6c11c — 2025-12-04 00:39:101.03 (n/a)1.03 (n/a)1.03 (n/a)1.02 (n/a)0.00 (n/a)24594.30 (n/a)24542.14 (n/a)24551.90 (n/a)24495.50 (n/a)37.80 (n/a)

gemm_2048x2048x2048_64x64x64_2_cols_0_bcolmaj_0_ccolmaj_0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
a4b6ffe — 2026-01-05 19:24:051.04 (+0.28%)1.03 (-0.15%)1.03 (+0.05%)1.02 (-0.52%)0.01 (+42.72%)24719.80 (+0.52%)24512.42 (+0.16%)24529.20 (-0.05%)24092.60 (-0.28%)252.92 (+43.15%)713.08 (+0.28%)700.92 (-0.15%)700.38 (+0.05%)694.98 (-0.52%)7.30 (+42.71%)
50ad37c — 2025-12-29 20:35:531.04 (n/a)1.03 (n/a)1.03 (n/a)1.02 (n/a)0.01 (n/a)24592.30 (n/a)24473.70 (n/a)24540.90 (n/a)24160.80 (n/a)176.68 (n/a)711.06 (n/a)702.00 (n/a)700.05 (n/a)698.59 (n/a)5.12 (n/a)

gemm_2048x2048x2048_64x64x64_2_cols_0_bcolmaj_1_ccolmaj_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.99 (+1.69%)0.97 (+0.29%)0.97 (-0.04%)0.97 (-0.15%)0.01 (+434.92%)26025.30 (+0.15%)25839.34 (-0.29%)25916.70 (+0.04%)25446.00 (-1.67%)202.05 (+425.44%)
0a6c11c — 2025-12-04 00:39:100.97 (n/a)0.97 (n/a)0.97 (n/a)0.97 (n/a)0.00 (n/a)25986.90 (n/a)25913.52 (n/a)25905.70 (n/a)25877.00 (n/a)38.45 (n/a)

gemm_2048x2048x2048_64x64x64_2_cols_0_bcolmaj_1_ccolmaj_0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
a4b6ffe — 2026-01-05 19:24:050.97 (-0.01%)0.97 (+0.03%)0.97 (+0.04%)0.97 (+0.00%)0.00 (-6.43%)25927.90 (-0.00%)25903.20 (-0.03%)25898.60 (-0.04%)25892.80 (+0.01%)14.05 (-6.56%)663.50 (-0.01%)663.23 (+0.03%)663.35 (+0.04%)662.60 (+0.00%)0.36 (-6.43%)
50ad37c — 2025-12-29 20:35:530.97 (n/a)0.97 (n/a)0.97 (n/a)0.97 (n/a)0.00 (n/a)25928.10 (n/a)25910.14 (n/a)25908.10 (n/a)25891.20 (n/a)15.04 (n/a)663.54 (n/a)663.06 (n/a)663.11 (n/a)662.60 (n/a)0.38 (n/a)

gemm_2048x2048x2048_64x64x64_2_cols_1_bcolmaj_0_ccolmaj_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.92 (-1.02%)0.92 (-0.66%)0.92 (-0.10%)0.91 (-0.78%)0.01 (-4.54%)27751.40 (+0.79%)27480.74 (+0.66%)27354.80 (+0.10%)27320.50 (+1.03%)182.69 (-2.84%)
0a6c11c — 2025-12-04 00:39:100.93 (n/a)0.92 (n/a)0.92 (n/a)0.91 (n/a)0.01 (n/a)27534.80 (n/a)27299.32 (n/a)27326.70 (n/a)27042.30 (n/a)188.04 (n/a)

gemm_2048x2048x2048_64x64x64_2_cols_1_bcolmaj_0_ccolmaj_0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
a4b6ffe — 2026-01-05 19:24:050.92 (+0.11%)0.91 (-0.12%)0.91 (-0.21%)0.91 (-0.24%)0.00 (+30.34%)27723.50 (+0.24%)27548.94 (+0.12%)27522.90 (+0.21%)27403.60 (-0.11%)124.26 (+30.58%)626.92 (+0.11%)623.62 (-0.12%)624.20 (-0.21%)619.69 (-0.24%)2.81 (+30.34%)
50ad37c — 2025-12-29 20:35:530.92 (n/a)0.91 (n/a)0.92 (n/a)0.91 (n/a)0.00 (n/a)27657.10 (n/a)27515.58 (n/a)27464.70 (n/a)27434.60 (n/a)95.16 (n/a)626.21 (n/a)624.37 (n/a)625.52 (n/a)621.17 (n/a)2.16 (n/a)

gemm_2048x2048x2048_64x64x64_2cols_bcolmaj

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
4a70412 — 2026-02-09 15:12:370.91 (-2.07%)0.91 (-0.61%)0.91 (-0.59%)0.91 (+0.27%)0.00 (-88.77%)27751.90 (-0.27%)27699.98 (+0.60%)27684.20 (+0.60%)27678.00 (+2.11%)31.20 (-88.55%)620.70 (-2.07%)620.21 (-0.61%)620.57 (-0.59%)619.05 (+0.27%)0.70 (-88.77%)
1531781 — 2026-01-23 18:49:240.93 (n/a)0.91 (n/a)0.91 (n/a)0.90 (n/a)0.01 (n/a)27826.70 (n/a)27533.84 (n/a)27519.90 (n/a)27105.20 (n/a)272.35 (n/a)633.82 (n/a)624.00 (n/a)624.27 (n/a)617.39 (n/a)6.21 (n/a)

gemm_2048x2048x2048_64x64x64_8_cols_0_bcolmaj_0_ccolmaj_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)GFLOP/s (max)GFLOP/s (mean)GFLOP/s (median)GFLOP/s (min)GFLOP/s (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
31ab24e — 2025-11-22 02:01:563.65 (-1.06%)3.59 (+0.83%)3.63 (-0.26%)3.45 (+2.66%)0.07 (-44.71%)2678.08 (-1.06%)2632.13 (+0.83%)2658.60 (-0.26%)2528.31 (+2.66%)53.86 (-44.71%)6795.00 (-2.59%)6529.80 (-0.92%)6462.00 (+0.26%)6415.00 (+1.07%)137.14 (-45.27%)
d9a64c5 — 2025-11-21 20:49:353.69 (n/a)3.56 (n/a)3.64 (n/a)3.36 (n/a)0.13 (n/a)2706.77 (n/a)2610.50 (n/a)2665.61 (n/a)2462.71 (n/a)97.41 (n/a)6976.00 (n/a)6590.40 (n/a)6445.00 (n/a)6347.00 (n/a)250.58 (n/a)

gemm_2048x2048x2048_64x64x64_8_cols_0_bcolmaj_0_ccolmaj_0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
d48746f — 2026-01-12 22:17:103.98 (+0.20%)3.84 (-2.95%)3.76 (-5.27%)3.72 (-4.97%)0.13 (+473.97%)6760.00 (+5.23%)6559.50 (+3.13%)6697.10 (+5.56%)6321.00 (-0.20%)218.68 (+499.40%)2717.91 (+0.20%)2621.44 (-2.95%)2565.27 (-5.27%)2541.40 (-4.97%)88.37 (+473.97%)
331dcca — 2026-01-08 18:34:103.97 (n/a)3.96 (n/a)3.97 (n/a)3.92 (n/a)0.02 (n/a)6423.80 (n/a)6360.32 (n/a)6344.30 (n/a)6333.90 (n/a)36.48 (n/a)2712.39 (n/a)2701.17 (n/a)2707.91 (n/a)2674.40 (n/a)15.40 (n/a)

gemm_2048x2048x2048_64x64x64_8_cols_0_bcolmaj_1_ccolmaj_0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
d48746f — 2026-01-12 22:17:103.75 (+0.28%)3.70 (+1.25%)3.71 (-0.21%)3.65 (+4.77%)0.04 (-66.90%)6901.80 (-4.56%)6795.46 (-1.30%)6790.20 (+0.21%)6715.90 (-0.28%)67.05 (-68.40%)2558.08 (+0.28%)2528.33 (+1.25%)2530.10 (-0.21%)2489.18 (+4.77%)24.81 (-66.90%)
331dcca — 2026-01-08 18:34:103.74 (n/a)3.66 (n/a)3.71 (n/a)3.48 (n/a)0.11 (n/a)7231.20 (n/a)6885.18 (n/a)6775.70 (n/a)6734.90 (n/a)212.21 (n/a)2550.87 (n/a)2497.04 (n/a)2535.50 (n/a)2375.78 (n/a)74.96 (n/a)

gemm_2048x2048x2048_64x64x64_8_cols_1_bcolmaj_0_ccolmaj_0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
d48746f — 2026-01-12 22:17:103.62 (-0.77%)3.54 (+0.14%)3.60 (-0.04%)3.44 (+2.17%)0.09 (-25.77%)7307.40 (-2.13%)7103.46 (-0.18%)6982.50 (+0.04%)6948.20 (+0.77%)182.95 (-26.55%)2472.55 (-0.77%)2419.80 (+0.14%)2460.41 (-0.04%)2351.04 (+2.17%)61.76 (-25.77%)
331dcca — 2026-01-08 18:34:103.65 (n/a)3.54 (n/a)3.61 (n/a)3.37 (n/a)0.12 (n/a)7466.20 (n/a)7116.40 (n/a)6979.60 (n/a)6895.10 (n/a)249.09 (n/a)2491.61 (n/a)2416.45 (n/a)2461.45 (n/a)2301.00 (n/a)83.20 (n/a)

gemm_2048x2048x2048_64x64x64_8cols_bcolmaj_ccolmaj

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
4a70412 — 2026-02-09 15:12:373.36 (-3.15%)3.25 (-4.88%)3.24 (-5.94%)3.16 (-3.99%)0.09 (+19.78%)7968.90 (+4.15%)7748.14 (+5.15%)7761.50 (+6.31%)7500.20 (+3.26%)209.23 (+28.73%)2290.60 (-3.15%)2218.59 (-4.88%)2213.48 (-5.94%)2155.86 (-3.99%)60.07 (+19.78%)
1531781 — 2026-01-23 18:49:243.46 (n/a)3.42 (n/a)3.45 (n/a)3.29 (n/a)0.07 (n/a)7651.20 (n/a)7368.34 (n/a)7300.80 (n/a)7263.70 (n/a)162.53 (n/a)2365.17 (n/a)2332.46 (n/a)2353.14 (n/a)2245.39 (n/a)50.15 (n/a)

gemm_384x1536x1792_32x48x64_4cols_bcolmaj

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
4a70412 — 2026-02-09 15:12:374.29 (-1.45%)4.00 (-5.63%)4.17 (-3.68%)3.50 (-8.97%)0.34 (+54.90%)2301.50 (+9.86%)2026.98 (+6.37%)1933.10 (+3.82%)1879.20 (+1.47%)181.43 (+71.35%)1124.90 (-1.45%)1049.26 (-5.63%)1093.55 (-3.68%)918.52 (-8.97%)89.09 (+54.90%)
1531781 — 2026-01-23 18:49:244.35 (n/a)4.24 (n/a)4.33 (n/a)3.85 (n/a)0.22 (n/a)2094.90 (n/a)1905.64 (n/a)1862.00 (n/a)1852.00 (n/a)105.88 (n/a)1141.45 (n/a)1111.86 (n/a)1135.29 (n/a)1009.08 (n/a)57.51 (n/a)

gemm_896x1792x640_32x64x80_8cols_ccolmaj

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
4a70412 — 2026-02-09 15:12:376.37 (-6.36%)5.72 (-11.09%)6.17 (-7.94%)4.80 (-9.95%)0.75 (+21.15%)1385.00 (+11.06%)1179.90 (+13.16%)1078.60 (+8.63%)1044.80 (+6.80%)162.42 (+41.74%)1967.18 (-6.36%)1767.32 (-11.09%)1905.43 (-7.94%)1483.93 (-9.95%)231.19 (+21.15%)
1531781 — 2026-01-23 18:49:246.80 (n/a)6.43 (n/a)6.70 (n/a)5.33 (n/a)0.62 (n/a)1247.10 (n/a)1042.70 (n/a)992.90 (n/a)978.30 (n/a)114.59 (n/a)2100.72 (n/a)1987.76 (n/a)2069.85 (n/a)1647.93 (n/a)190.83 (n/a)

layer_norm_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.07 (-41.54%)0.05 (-40.25%)0.05 (-36.40%)0.03 (-50.08%)0.01 (-40.23%)239.80 (+100.33%)164.60 (+68.92%)158.10 (+57.31%)122.30 (+71.05%)46.31 (+104.30%)
1531781 — 2026-01-23 18:49:240.11 (n/a)0.09 (n/a)0.08 (n/a)0.07 (n/a)0.02 (n/a)119.70 (n/a)97.44 (n/a)100.50 (n/a)71.50 (n/a)22.67 (n/a)

layer_norm_1_cols_2_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.07 (-37.40%)0.06 (-32.82%)0.06 (-34.53%)0.05 (-16.39%)0.01 (-59.42%)165.20 (+19.62%)142.58 (+44.34%)148.30 (+52.73%)122.10 (+59.82%)19.06 (-23.38%)
1531781 — 2026-01-23 18:49:240.11 (n/a)0.09 (n/a)0.08 (n/a)0.06 (n/a)0.02 (n/a)138.10 (n/a)98.78 (n/a)97.10 (n/a)76.40 (n/a)24.88 (n/a)

layer_norm_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.07 (-44.66%)0.06 (-33.66%)0.06 (-34.57%)0.05 (-18.02%)0.01 (-64.13%)170.90 (+21.98%)142.80 (+45.15%)144.20 (+52.75%)123.00 (+80.62%)19.89 (-23.83%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.09 (n/a)0.09 (n/a)0.06 (n/a)0.02 (n/a)140.10 (n/a)98.38 (n/a)94.40 (n/a)68.10 (n/a)26.11 (n/a)

layer_norm_2_cols_2_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.05 (-60.59%)0.04 (-53.90%)0.05 (-48.39%)0.03 (-52.60%)0.01 (-63.78%)275.10 (+110.97%)195.94 (+113.21%)181.20 (+93.80%)157.40 (+153.87%)48.97 (+91.40%)
1531781 — 2026-01-23 18:49:240.13 (n/a)0.09 (n/a)0.09 (n/a)0.06 (n/a)0.03 (n/a)130.40 (n/a)91.90 (n/a)93.50 (n/a)62.00 (n/a)25.59 (n/a)

layer_norm_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.07 (-41.86%)0.05 (-43.39%)0.05 (-50.35%)0.04 (-30.68%)0.01 (-49.14%)203.40 (+44.26%)158.08 (+72.69%)156.90 (+101.41%)125.60 (+72.05%)33.79 (+18.87%)
1531781 — 2026-01-23 18:49:240.11 (n/a)0.09 (n/a)0.11 (n/a)0.06 (n/a)0.02 (n/a)141.00 (n/a)91.54 (n/a)77.90 (n/a)73.00 (n/a)28.43 (n/a)

layer_norm_4_cols_2_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.06 (-49.79%)0.05 (-41.24%)0.05 (-46.40%)0.04 (-19.56%)0.00 (-81.76%)184.50 (+24.33%)164.32 (+59.32%)163.80 (+86.56%)146.90 (+99.32%)14.50 (-55.09%)
1531781 — 2026-01-23 18:49:240.11 (n/a)0.09 (n/a)0.09 (n/a)0.06 (n/a)0.02 (n/a)148.40 (n/a)103.14 (n/a)87.80 (n/a)73.70 (n/a)32.30 (n/a)

layer_norm_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.05 (-59.33%)0.04 (-55.97%)0.04 (-59.67%)0.04 (-24.05%)0.00 (-86.53%)202.70 (+31.71%)188.84 (+108.99%)191.90 (+147.93%)165.50 (+145.91%)14.60 (-59.29%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.10 (n/a)0.11 (n/a)0.05 (n/a)0.03 (n/a)153.90 (n/a)90.36 (n/a)77.40 (n/a)67.30 (n/a)35.86 (n/a)

layer_norm_8_cols_2_channels_2048_tile_128

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.05 (-48.40%)0.04 (-50.24%)0.04 (-51.87%)0.03 (-48.78%)0.01 (-49.99%)273.90 (+95.22%)213.76 (+100.71%)202.90 (+107.89%)179.10 (+93.83%)37.52 (+89.15%)
1531781 — 2026-01-23 18:49:240.09 (n/a)0.08 (n/a)0.08 (n/a)0.06 (n/a)0.01 (n/a)140.30 (n/a)106.50 (n/a)97.60 (n/a)92.40 (n/a)19.84 (n/a)

matrix_vector_mul_128x128_32_1col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:490.37 (+17.13%)0.30 (+24.87%)0.30 (+38.03%)0.23 (+12.47%)0.06 (+35.23%)
0a6c11c — 2025-12-04 00:39:100.32 (n/a)0.24 (n/a)0.22 (n/a)0.20 (n/a)0.04 (n/a)

matrix_vector_mul_128x128_32_1col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:100.26 (-1.81%)0.24 (+4.06%)0.24 (+0.52%)0.21 (+16.97%)0.02 (-48.16%)0.26 (-1.81%)0.24 (+4.06%)0.24 (+0.52%)0.20 (+16.97%)0.02 (-48.16%)
a4b6ffe — 2026-01-05 19:24:050.26 (n/a)0.23 (n/a)0.24 (n/a)0.18 (n/a)0.04 (n/a)0.26 (n/a)0.23 (n/a)0.24 (n/a)0.17 (n/a)0.04 (n/a)

matrix_vector_mul_128x128_32tsi_128tso_1col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
4a70412 — 2026-02-09 15:12:370.25 (-43.29%)0.20 (-43.17%)0.21 (-39.84%)0.13 (-39.85%)0.05 (-43.34%)0.25 (-43.29%)0.19 (-43.17%)0.21 (-39.84%)0.13 (-39.85%)0.05 (-43.34%)
1531781 — 2026-01-23 18:49:240.45 (n/a)0.35 (n/a)0.35 (n/a)0.22 (n/a)0.08 (n/a)0.44 (n/a)0.34 (n/a)0.34 (n/a)0.22 (n/a)0.08 (n/a)

matrix_vector_mul_2048x8192_1_1col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:4913.57 (+3.47%)13.15 (+1.09%)13.10 (+0.16%)12.52 (-1.58%)0.37 (+153.19%)
0a6c11c — 2025-12-04 00:39:1013.11 (n/a)13.01 (n/a)13.07 (n/a)12.72 (n/a)0.15 (n/a)

matrix_vector_mul_2048x8192_1_1col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:1013.43 (-0.07%)13.30 (+4.62%)13.27 (+1.29%)13.16 (+16.70%)0.12 (-86.69%)13.42 (-0.07%)13.29 (+4.62%)13.26 (+1.29%)13.16 (+16.70%)0.12 (-86.69%)
a4b6ffe — 2026-01-05 19:24:0513.44 (n/a)12.71 (n/a)13.10 (n/a)11.28 (n/a)0.87 (n/a)13.43 (n/a)12.70 (n/a)13.09 (n/a)11.27 (n/a)0.87 (n/a)

matrix_vector_mul_2048x8192_1_2col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:4925.06 (+2.47%)23.71 (+1.81%)24.13 (+0.50%)20.90 (+5.41%)1.48 (-15.60%)
0a6c11c — 2025-12-04 00:39:1024.46 (n/a)23.29 (n/a)24.01 (n/a)19.83 (n/a)1.75 (n/a)

matrix_vector_mul_2048x8192_1_2col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:1024.46 (-0.60%)21.07 (-8.03%)19.75 (-17.83%)18.08 (+1.18%)3.03 (+7.07%)24.44 (-0.60%)21.06 (-8.03%)19.74 (-17.83%)18.07 (+1.18%)3.03 (+7.07%)
a4b6ffe — 2026-01-05 19:24:0524.60 (n/a)22.91 (n/a)24.04 (n/a)17.87 (n/a)2.83 (n/a)24.59 (n/a)22.90 (n/a)24.02 (n/a)17.86 (n/a)2.83 (n/a)

matrix_vector_mul_2048x8192_1_4col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:4942.05 (+1.54%)40.50 (+12.60%)40.49 (+2.36%)38.98 (+97.53%)0.97 (-88.09%)
0a6c11c — 2025-12-04 00:39:1041.41 (n/a)35.97 (n/a)39.55 (n/a)19.73 (n/a)8.15 (n/a)

matrix_vector_mul_2048x8192_1_4col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:1042.58 (+1.51%)41.37 (+14.29%)41.25 (+4.89%)40.33 (+91.42%)0.92 (-89.18%)42.56 (+1.51%)41.34 (+14.29%)41.23 (+4.89%)40.30 (+91.42%)0.92 (-89.18%)
a4b6ffe — 2026-01-05 19:24:0541.95 (n/a)36.20 (n/a)39.33 (n/a)21.07 (n/a)8.54 (n/a)41.92 (n/a)36.17 (n/a)39.31 (n/a)21.05 (n/a)8.53 (n/a)

matrix_vector_mul_2048x8192_1_8col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:4946.62 (+3.11%)44.82 (+0.17%)44.78 (+0.17%)42.60 (-3.80%)1.60 (+367.72%)
0a6c11c — 2025-12-04 00:39:1045.21 (n/a)44.75 (n/a)44.70 (n/a)44.29 (n/a)0.34 (n/a)

matrix_vector_mul_2048x8192_1_8col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:1046.04 (+4.14%)45.06 (+3.47%)44.55 (+2.76%)44.30 (+2.84%)0.87 (+66.58%)46.01 (+4.14%)45.03 (+3.47%)44.52 (+2.76%)44.27 (+2.84%)0.87 (+66.58%)
a4b6ffe — 2026-01-05 19:24:0544.21 (n/a)43.55 (n/a)43.35 (n/a)43.07 (n/a)0.52 (n/a)44.18 (n/a)43.52 (n/a)43.32 (n/a)43.05 (n/a)0.52 (n/a)

matrix_vector_mul_2048x8192_1tsi_1024tso_2col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
4a70412 — 2026-02-09 15:12:3725.11 (+1.24%)24.48 (-0.44%)24.19 (-1.44%)24.16 (-0.97%)0.44 (+147.51%)25.10 (+1.24%)24.47 (-0.44%)24.18 (-1.44%)24.15 (-0.97%)0.44 (+147.50%)
1531781 — 2026-01-23 18:49:2424.80 (n/a)24.59 (n/a)24.55 (n/a)24.40 (n/a)0.18 (n/a)24.79 (n/a)24.58 (n/a)24.53 (n/a)24.38 (n/a)0.18 (n/a)

matrix_vector_mul_2048x8192_1tsi_2048tso_1col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
4a70412 — 2026-02-09 15:12:3713.51 (-0.19%)13.02 (-2.43%)12.97 (-3.54%)12.39 (-5.40%)0.45 (+118.80%)13.51 (-0.19%)13.01 (-2.43%)12.96 (-3.54%)12.39 (-5.40%)0.45 (+118.81%)
1531781 — 2026-01-23 18:49:2413.54 (n/a)13.34 (n/a)13.45 (n/a)13.10 (n/a)0.21 (n/a)13.53 (n/a)13.33 (n/a)13.44 (n/a)13.09 (n/a)0.21 (n/a)

matrix_vector_mul_2048x8192_1tsi_256tso_8col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
4a70412 — 2026-02-09 15:12:3748.16 (+2.88%)45.39 (-1.13%)46.77 (+0.85%)41.42 (-7.58%)2.74 (+207.20%)48.13 (+2.88%)45.36 (-1.13%)46.74 (+0.85%)41.39 (-7.58%)2.74 (+207.19%)
1531781 — 2026-01-23 18:49:2446.81 (n/a)45.91 (n/a)46.38 (n/a)44.81 (n/a)0.89 (n/a)46.79 (n/a)45.88 (n/a)46.35 (n/a)44.79 (n/a)0.89 (n/a)

matrix_vector_mul_2048x8192_1tsi_512tso_4col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
4a70412 — 2026-02-09 15:12:3742.95 (+3.51%)40.28 (-1.31%)39.92 (-2.94%)38.31 (-4.11%)1.77 (+171.59%)42.92 (+3.51%)40.25 (-1.31%)39.89 (-2.94%)38.29 (-4.11%)1.77 (+171.59%)
1531781 — 2026-01-23 18:49:2441.49 (n/a)40.81 (n/a)41.13 (n/a)39.95 (n/a)0.65 (n/a)41.46 (n/a)40.79 (n/a)41.10 (n/a)39.93 (n/a)0.65 (n/a)

matrix_vector_mul_8192x2048_4_1col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:4913.29 (+1.63%)13.19 (+6.08%)13.19 (+1.17%)13.09 (+26.86%)0.08 (-92.94%)
0a6c11c — 2025-12-04 00:39:1013.08 (n/a)12.44 (n/a)13.03 (n/a)10.32 (n/a)1.07 (n/a)

matrix_vector_mul_8192x2048_4_1col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:1013.68 (+3.42%)13.30 (+8.54%)13.38 (+7.25%)12.72 (+14.96%)0.35 (-66.09%)13.67 (+3.42%)13.29 (+8.54%)13.38 (+7.25%)12.71 (+14.96%)0.35 (-66.09%)
a4b6ffe — 2026-01-05 19:24:0513.23 (n/a)12.25 (n/a)12.48 (n/a)11.07 (n/a)1.04 (n/a)13.22 (n/a)12.24 (n/a)12.47 (n/a)11.06 (n/a)1.04 (n/a)

matrix_vector_mul_8192x2048_4_2col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:4925.58 (+4.24%)24.23 (+6.59%)24.03 (+0.29%)23.58 (+37.29%)0.70 (-74.97%)
0a6c11c — 2025-12-04 00:39:1024.54 (n/a)22.73 (n/a)23.96 (n/a)17.18 (n/a)2.79 (n/a)

matrix_vector_mul_8192x2048_4_2col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:1025.08 (+1.32%)24.68 (+13.42%)24.61 (+3.26%)24.47 (+48.32%)0.24 (-93.56%)25.07 (+1.32%)24.67 (+13.42%)24.59 (+3.26%)24.46 (+48.32%)0.24 (-93.56%)
a4b6ffe — 2026-01-05 19:24:0524.76 (n/a)21.76 (n/a)23.83 (n/a)16.50 (n/a)3.74 (n/a)24.74 (n/a)21.75 (n/a)23.82 (n/a)16.49 (n/a)3.73 (n/a)

matrix_vector_mul_8192x2048_4_4col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:4940.94 (-2.19%)39.74 (-1.10%)39.76 (-0.60%)38.17 (-2.14%)1.04 (+4.39%)
0a6c11c — 2025-12-04 00:39:1041.86 (n/a)40.18 (n/a)40.00 (n/a)39.00 (n/a)1.00 (n/a)

matrix_vector_mul_8192x2048_4_4col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:1040.74 (-1.09%)37.16 (-7.15%)40.25 (+1.47%)24.51 (-37.44%)7.07 (+736.13%)40.72 (-1.09%)37.13 (-7.15%)40.23 (+1.47%)24.50 (-37.44%)7.07 (+736.13%)
a4b6ffe — 2026-01-05 19:24:0541.19 (n/a)40.02 (n/a)39.67 (n/a)39.18 (n/a)0.85 (n/a)41.17 (n/a)39.99 (n/a)39.64 (n/a)39.16 (n/a)0.85 (n/a)

matrix_vector_mul_8192x2048_4_8col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:4946.71 (-1.04%)44.91 (-2.34%)44.71 (-2.98%)43.80 (-1.44%)0.96 (+6.93%)
0a6c11c — 2025-12-04 00:39:1047.20 (n/a)45.98 (n/a)46.08 (n/a)44.44 (n/a)0.90 (n/a)

matrix_vector_mul_8192x2048_4_8col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:1046.28 (+5.68%)44.89 (+14.59%)44.74 (+4.26%)44.20 (+62.80%)0.84 (-87.96%)46.25 (+5.68%)44.86 (+14.59%)44.71 (+4.26%)44.17 (+62.80%)0.84 (-87.96%)
a4b6ffe — 2026-01-05 19:24:0543.79 (n/a)39.17 (n/a)42.91 (n/a)27.15 (n/a)6.98 (n/a)43.77 (n/a)39.15 (n/a)42.88 (n/a)27.13 (n/a)6.98 (n/a)

matrix_vector_mul_8192x2048_4tsi_1024tso_1col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
4a70412 — 2026-02-09 15:12:3713.45 (-0.09%)13.14 (-1.38%)13.23 (-0.70%)12.49 (-5.35%)0.38 (+236.74%)13.44 (-0.09%)13.13 (-1.38%)13.22 (-0.70%)12.49 (-5.35%)0.38 (+236.74%)
1531781 — 2026-01-23 18:49:2413.46 (n/a)13.33 (n/a)13.32 (n/a)13.20 (n/a)0.11 (n/a)13.45 (n/a)13.32 (n/a)13.31 (n/a)13.19 (n/a)0.11 (n/a)

matrix_vector_mul_8192x2048_4tsi_1024tso_2col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
4a70412 — 2026-02-09 15:12:3725.12 (-0.58%)24.34 (-1.66%)24.41 (-0.83%)23.62 (-3.76%)0.65 (+118.03%)25.10 (-0.58%)24.33 (-1.66%)24.39 (-0.83%)23.61 (-3.76%)0.65 (+118.02%)
1531781 — 2026-01-23 18:49:2425.27 (n/a)24.75 (n/a)24.61 (n/a)24.55 (n/a)0.30 (n/a)25.25 (n/a)24.74 (n/a)24.59 (n/a)24.53 (n/a)0.30 (n/a)

matrix_vector_mul_8192x2048_4tsi_1024tso_4col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
4a70412 — 2026-02-09 15:12:3743.03 (+5.30%)39.85 (-0.40%)39.10 (-3.30%)37.52 (-3.45%)2.30 (+159.61%)43.00 (+5.30%)39.82 (-0.40%)39.07 (-3.30%)37.50 (-3.45%)2.30 (+159.61%)
1531781 — 2026-01-23 18:49:2440.86 (n/a)40.01 (n/a)40.43 (n/a)38.86 (n/a)0.89 (n/a)40.84 (n/a)39.98 (n/a)40.41 (n/a)38.84 (n/a)0.89 (n/a)

matrix_vector_mul_8192x2048_4tsi_1024tso_8col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
4a70412 — 2026-02-09 15:12:3744.44 (-3.57%)43.26 (-4.10%)43.13 (-5.19%)42.24 (-3.77%)0.79 (-17.82%)44.41 (-3.57%)43.23 (-4.10%)43.10 (-5.19%)42.21 (-3.77%)0.79 (-17.82%)
1531781 — 2026-01-23 18:49:2446.08 (n/a)45.11 (n/a)45.49 (n/a)43.89 (n/a)0.97 (n/a)46.05 (n/a)45.08 (n/a)45.46 (n/a)43.87 (n/a)0.97 (n/a)

mem_copy_16_cores_2_chans_2048_tile_128_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.07 (+22.03%)0.06 (+27.86%)0.06 (+35.49%)0.05 (+15.00%)0.01 (+49.69%)173.30 (-13.00%)137.94 (-21.32%)133.60 (-26.19%)123.40 (-18.06%)18.40 (+8.53%)
0a6c11c — 2025-12-04 00:39:100.05 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.00 (n/a)199.20 (n/a)175.32 (n/a)181.00 (n/a)150.60 (n/a)16.96 (n/a)

mem_copy_16_cores_2_chans_2048_tile_128_False0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.05 (-43.39%)0.05 (-38.97%)0.05 (-50.10%)0.05 (+9.52%)0.00 (-81.99%)178.30 (-8.70%)166.04 (+48.75%)173.60 (+100.46%)149.50 (+76.71%)13.74 (-71.16%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.08 (n/a)0.09 (n/a)0.04 (n/a)0.02 (n/a)195.30 (n/a)111.62 (n/a)86.60 (n/a)84.60 (n/a)47.65 (n/a)

mem_copy_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
003b49d — 2025-11-14 19:30:540.05 (-16.99%)0.05 (-12.47%)0.05 (-7.32%)0.04 (-11.17%)0.00 (-29.61%)197.00 (+12.57%)171.00 (+13.70%)164.00 (+7.89%)153.00 (+20.47%)17.49 (-4.32%)
0999a9a — 2025-11-13 19:24:080.06 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)175.00 (n/a)150.40 (n/a)152.00 (n/a)127.00 (n/a)18.28 (n/a)

mem_copy_1_cols_2_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
003b49d — 2025-11-14 19:30:540.06 (+5.56%)0.05 (-1.59%)0.05 (-1.39%)0.03 (-14.07%)0.01 (+25.03%)263.00 (+16.37%)166.00 (+4.53%)144.00 (+1.41%)126.00 (-5.26%)49.38 (+43.17%)
0999a9a — 2025-11-13 19:24:080.06 (n/a)0.05 (n/a)0.05 (n/a)0.03 (n/a)0.01 (n/a)226.00 (n/a)158.80 (n/a)142.00 (n/a)133.00 (n/a)34.49 (n/a)

mem_copy_1_cores_1_chans_2048_tile_2048_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.08 (+10.93%)0.07 (+7.79%)0.07 (+3.55%)0.05 (-1.73%)0.01 (+36.73%)150.20 (+1.76%)121.68 (-6.36%)125.60 (-3.38%)97.90 (-9.85%)18.56 (+21.98%)
0a6c11c — 2025-12-04 00:39:100.08 (n/a)0.06 (n/a)0.06 (n/a)0.06 (n/a)0.01 (n/a)147.60 (n/a)129.94 (n/a)130.00 (n/a)108.60 (n/a)15.21 (n/a)

mem_copy_1_cores_1_chans_2048_tile_2048_False0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.08 (-20.57%)0.07 (-23.31%)0.07 (-26.98%)0.05 (-27.47%)0.01 (-26.19%)162.80 (+37.85%)125.66 (+30.19%)115.60 (+36.97%)103.70 (+25.85%)23.74 (+29.80%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.09 (n/a)0.10 (n/a)0.07 (n/a)0.02 (n/a)118.10 (n/a)96.52 (n/a)84.40 (n/a)82.40 (n/a)18.29 (n/a)

mem_copy_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
003b49d — 2025-11-14 19:30:540.06 (-15.32%)0.05 (-17.09%)0.05 (-23.81%)0.04 (-15.54%)0.01 (-18.18%)193.00 (+18.40%)160.80 (+20.36%)168.00 (+31.25%)124.00 (+18.10%)22.92 (+11.39%)
0999a9a — 2025-11-13 19:24:080.07 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)163.00 (n/a)133.60 (n/a)128.00 (n/a)105.00 (n/a)20.58 (n/a)

mem_copy_2_cols_2_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
003b49d — 2025-11-14 19:30:540.06 (-23.02%)0.05 (-6.98%)0.05 (-2.13%)0.04 (-15.31%)0.01 (-33.29%)196.00 (+18.07%)147.00 (+6.52%)141.00 (+2.17%)126.00 (+29.90%)25.27 (+8.13%)
0999a9a — 2025-11-13 19:24:080.08 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)166.00 (n/a)138.00 (n/a)138.00 (n/a)97.00 (n/a)23.37 (n/a)

mem_copy_2_cores_1_chans_2048_tile_1024_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.10 (+31.37%)0.08 (+29.12%)0.08 (+29.06%)0.07 (+41.80%)0.01 (+0.23%)109.90 (-29.46%)98.30 (-23.38%)100.80 (-22.52%)79.40 (-23.87%)10.54 (-46.17%)
0a6c11c — 2025-12-04 00:39:100.08 (n/a)0.07 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)155.80 (n/a)128.30 (n/a)130.10 (n/a)104.30 (n/a)19.58 (n/a)

mem_copy_2_cores_1_chans_2048_tile_1024_False0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.07 (-40.98%)0.06 (-31.09%)0.06 (-31.50%)0.06 (-16.86%)0.00 (-82.93%)140.90 (+20.32%)127.56 (+38.35%)126.80 (+45.91%)119.20 (+69.32%)8.16 (-64.99%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.09 (n/a)0.09 (n/a)0.07 (n/a)0.02 (n/a)117.10 (n/a)92.20 (n/a)86.90 (n/a)70.40 (n/a)23.32 (n/a)

mem_copy_2_cores_2_chans_2048_tile_1024_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.10 (+25.02%)0.07 (+10.01%)0.07 (+7.53%)0.05 (-1.99%)0.01 (+77.08%)150.10 (+2.04%)114.60 (-7.22%)111.70 (-6.99%)83.90 (-20.02%)21.23 (+43.54%)
0a6c11c — 2025-12-04 00:39:100.08 (n/a)0.07 (n/a)0.07 (n/a)0.06 (n/a)0.01 (n/a)147.10 (n/a)123.52 (n/a)120.10 (n/a)104.90 (n/a)14.79 (n/a)

mem_copy_2_cores_2_chans_2048_tile_1024_False0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.07 (-25.82%)0.06 (-23.20%)0.07 (-26.46%)0.06 (-3.31%)0.01 (-61.85%)145.00 (+3.42%)128.28 (+26.19%)125.20 (+35.94%)111.80 (+34.86%)13.19 (-45.56%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.08 (n/a)0.09 (n/a)0.06 (n/a)0.02 (n/a)140.20 (n/a)101.66 (n/a)92.10 (n/a)82.90 (n/a)24.23 (n/a)

mem_copy_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
003b49d — 2025-11-14 19:30:540.07 (+11.97%)0.05 (+3.09%)0.05 (+2.10%)0.03 (-11.35%)0.01 (+41.65%)229.00 (+12.81%)156.00 (-0.51%)143.00 (-2.05%)117.00 (-10.69%)38.21 (+48.71%)
0999a9a — 2025-11-13 19:24:080.06 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)203.00 (n/a)156.80 (n/a)146.00 (n/a)131.00 (n/a)25.69 (n/a)

mem_copy_4_cols_2_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
003b49d — 2025-11-14 19:30:540.06 (+9.02%)0.05 (+5.59%)0.05 (+0.00%)0.04 (+3.63%)0.01 (+25.48%)193.00 (-3.50%)147.40 (-4.66%)147.00 (+0.00%)122.00 (-8.27%)25.24 (+7.48%)
0999a9a — 2025-11-13 19:24:080.06 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)200.00 (n/a)154.60 (n/a)147.00 (n/a)133.00 (n/a)23.48 (n/a)

mem_copy_4_cores_1_chans_2048_tile_512_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.09 (-3.98%)0.07 (+4.00%)0.07 (+7.04%)0.06 (+3.67%)0.01 (-15.76%)129.70 (-3.57%)113.30 (-4.50%)116.90 (-6.55%)90.40 (+4.15%)15.04 (-13.80%)
0a6c11c — 2025-12-04 00:39:100.09 (n/a)0.07 (n/a)0.07 (n/a)0.06 (n/a)0.01 (n/a)134.50 (n/a)118.64 (n/a)125.10 (n/a)86.80 (n/a)17.45 (n/a)

mem_copy_4_cores_1_chans_2048_tile_512_False0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.07 (-43.33%)0.06 (-39.38%)0.07 (-42.16%)0.05 (-20.37%)0.01 (-65.51%)161.20 (+25.64%)133.14 (+57.45%)123.70 (+72.77%)115.90 (+76.41%)20.21 (-23.49%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.10 (n/a)0.11 (n/a)0.06 (n/a)0.03 (n/a)128.30 (n/a)84.56 (n/a)71.60 (n/a)65.70 (n/a)26.42 (n/a)

mem_copy_4_cores_2_chans_2048_tile_512_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.08 (+9.57%)0.07 (+16.85%)0.07 (+19.58%)0.06 (+24.52%)0.01 (+1.05%)134.50 (-19.70%)116.64 (-14.92%)116.80 (-16.39%)98.00 (-8.75%)15.71 (-25.13%)
0a6c11c — 2025-12-04 00:39:100.08 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)167.50 (n/a)137.10 (n/a)139.70 (n/a)107.40 (n/a)20.99 (n/a)

mem_copy_4_cores_2_chans_2048_tile_512_False0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.09 (-14.11%)0.06 (-33.20%)0.06 (-42.87%)0.05 (-19.29%)0.01 (-13.06%)165.00 (+23.87%)139.38 (+49.90%)145.90 (+74.94%)95.60 (+16.44%)27.30 (+21.38%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.09 (n/a)0.10 (n/a)0.06 (n/a)0.02 (n/a)133.20 (n/a)92.98 (n/a)83.40 (n/a)82.10 (n/a)22.49 (n/a)

mem_copy_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
003b49d — 2025-11-14 19:30:540.05 (-13.48%)0.05 (-17.79%)0.05 (-15.33%)0.03 (-25.09%)0.01 (+6.67%)267.00 (+33.50%)176.80 (+24.51%)150.00 (+18.11%)141.00 (+15.57%)47.41 (+61.41%)
0999a9a — 2025-11-13 19:24:080.06 (n/a)0.06 (n/a)0.06 (n/a)0.04 (n/a)0.01 (n/a)200.00 (n/a)142.00 (n/a)127.00 (n/a)122.00 (n/a)29.37 (n/a)

mem_copy_8_cols_2_channels_2048_tile_128

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
003b49d — 2025-11-14 19:30:540.05 (+13.38%)0.05 (+23.85%)0.05 (+37.91%)0.05 (+30.72%)0.00 (-56.85%)166.00 (-23.50%)153.40 (-20.60%)153.00 (-27.49%)142.00 (-11.80%)7.63 (-70.60%)
0999a9a — 2025-11-13 19:24:080.05 (n/a)0.04 (n/a)0.04 (n/a)0.04 (n/a)0.01 (n/a)217.00 (n/a)193.20 (n/a)211.00 (n/a)161.00 (n/a)25.96 (n/a)

mem_copy_8_cores_1_chans_2048_tile_256_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.09 (+16.31%)0.08 (+15.34%)0.08 (+14.60%)0.07 (+19.30%)0.01 (-5.41%)117.30 (-16.15%)107.80 (-13.61%)108.50 (-12.78%)94.10 (-13.99%)8.29 (-32.01%)
0a6c11c — 2025-12-04 00:39:100.07 (n/a)0.07 (n/a)0.07 (n/a)0.06 (n/a)0.01 (n/a)139.90 (n/a)124.78 (n/a)124.40 (n/a)109.40 (n/a)12.20 (n/a)

mem_copy_8_cores_1_chans_2048_tile_256_False0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.07 (-26.49%)0.06 (-24.77%)0.07 (-27.07%)0.05 (-22.09%)0.01 (-37.36%)168.60 (+28.41%)131.68 (+31.97%)121.60 (+37.09%)118.30 (+36.13%)21.09 (+11.58%)
1531781 — 2026-01-23 18:49:240.09 (n/a)0.08 (n/a)0.09 (n/a)0.06 (n/a)0.01 (n/a)131.30 (n/a)99.78 (n/a)88.70 (n/a)86.90 (n/a)18.90 (n/a)

mem_copy_8_cores_2_chans_2048_tile_256_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.09 (+12.61%)0.07 (-2.24%)0.07 (-8.85%)0.06 (+5.13%)0.01 (+35.50%)136.50 (-4.88%)120.86 (+2.96%)125.10 (+9.64%)91.60 (-11.15%)15.72 (+10.46%)
0a6c11c — 2025-12-04 00:39:100.08 (n/a)0.07 (n/a)0.07 (n/a)0.06 (n/a)0.01 (n/a)143.50 (n/a)117.38 (n/a)114.10 (n/a)103.10 (n/a)14.23 (n/a)

mem_copy_8_cores_2_chans_2048_tile_256_False0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.05 (-47.56%)0.05 (-44.67%)0.05 (-46.50%)0.04 (-26.04%)0.00 (-75.75%)195.30 (+35.25%)170.04 (+74.04%)164.70 (+86.95%)152.60 (+90.75%)15.90 (-39.66%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.09 (n/a)0.09 (n/a)0.06 (n/a)0.02 (n/a)144.40 (n/a)97.70 (n/a)88.10 (n/a)80.00 (n/a)26.35 (n/a)

mha

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.18 (+0.16%)0.18 (+0.19%)0.18 (+0.12%)0.18 (+0.32%)0.00 (-52.24%)47253.90 (-0.32%)47213.70 (-0.19%)47210.90 (-0.12%)47167.70 (-0.16%)28.71 (-52.47%)
0a6c11c — 2025-12-04 00:39:100.18 (n/a)0.18 (n/a)0.18 (n/a)0.18 (n/a)0.00 (n/a)47405.00 (n/a)47301.72 (n/a)47265.90 (n/a)47241.80 (n/a)60.40 (n/a)

mha0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.18 (-0.03%)0.18 (-0.46%)0.18 (-0.61%)0.18 (-0.49%)0.00 (+97.39%)47601.20 (+0.49%)47452.78 (+0.47%)47522.50 (+0.62%)47084.80 (+0.03%)209.13 (+98.32%)
1531781 — 2026-01-23 18:49:240.18 (n/a)0.18 (n/a)0.18 (n/a)0.18 (n/a)0.00 (n/a)47367.50 (n/a)47232.68 (n/a)47232.00 (n/a)47073.00 (n/a)105.45 (n/a)

relu_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.08 (-32.23%)0.06 (-36.04%)0.05 (-46.99%)0.05 (-15.27%)0.01 (-56.55%)180.50 (+18.05%)149.80 (+47.41%)155.00 (+88.56%)107.80 (+47.67%)26.71 (-25.27%)
1531781 — 2026-01-23 18:49:240.11 (n/a)0.09 (n/a)0.10 (n/a)0.05 (n/a)0.03 (n/a)152.90 (n/a)101.62 (n/a)82.20 (n/a)73.00 (n/a)35.75 (n/a)

relu_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.07 (-29.20%)0.06 (-29.05%)0.06 (-37.50%)0.05 (-2.40%)0.01 (-52.92%)154.30 (+2.46%)138.20 (+36.48%)146.20 (+60.13%)117.30 (+41.33%)18.42 (-34.27%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.08 (n/a)0.09 (n/a)0.05 (n/a)0.02 (n/a)150.60 (n/a)101.26 (n/a)91.30 (n/a)83.00 (n/a)28.02 (n/a)

relu_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.08 (-18.01%)0.05 (-40.90%)0.05 (-49.47%)0.03 (-50.02%)0.02 (+14.15%)271.80 (+100.15%)173.56 (+80.42%)170.70 (+97.80%)101.30 (+21.90%)61.92 (+175.68%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.09 (n/a)0.09 (n/a)0.06 (n/a)0.02 (n/a)135.80 (n/a)96.20 (n/a)86.30 (n/a)83.10 (n/a)22.46 (n/a)

relu_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.09 (-20.72%)0.05 (-45.35%)0.05 (-50.05%)0.02 (-58.59%)0.02 (-2.61%)372.90 (+141.52%)192.02 (+105.77%)156.90 (+100.13%)95.60 (+26.12%)106.07 (+209.79%)
1531781 — 2026-01-23 18:49:240.11 (n/a)0.09 (n/a)0.10 (n/a)0.05 (n/a)0.02 (n/a)154.40 (n/a)93.32 (n/a)78.40 (n/a)75.80 (n/a)34.24 (n/a)

rms_norm_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.06 (-30.59%)0.05 (-37.57%)0.05 (-42.42%)0.04 (-29.46%)0.01 (-38.73%)186.40 (+41.75%)160.50 (+59.38%)160.50 (+73.70%)130.90 (+44.00%)21.31 (+22.84%)
1531781 — 2026-01-23 18:49:240.09 (n/a)0.08 (n/a)0.09 (n/a)0.06 (n/a)0.01 (n/a)131.50 (n/a)100.70 (n/a)92.40 (n/a)90.90 (n/a)17.35 (n/a)

rms_norm_1_cols_2_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.06 (-51.89%)0.04 (-60.81%)0.04 (-63.05%)0.02 (-56.47%)0.01 (-54.35%)333.50 (+129.68%)220.28 (+154.54%)195.00 (+170.46%)146.60 (+107.94%)72.00 (+119.48%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.10 (n/a)0.11 (n/a)0.06 (n/a)0.03 (n/a)145.20 (n/a)86.54 (n/a)72.10 (n/a)70.50 (n/a)32.80 (n/a)

rms_norm_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.06 (-37.64%)0.05 (-44.86%)0.05 (-49.38%)0.04 (-35.67%)0.01 (-48.79%)201.50 (+55.48%)176.82 (+80.02%)179.50 (+97.47%)144.00 (+60.36%)22.07 (+25.74%)
1531781 — 2026-01-23 18:49:240.09 (n/a)0.09 (n/a)0.09 (n/a)0.06 (n/a)0.01 (n/a)129.60 (n/a)98.22 (n/a)90.90 (n/a)89.80 (n/a)17.55 (n/a)

rms_norm_2_cols_2_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.06 (-55.72%)0.05 (-58.93%)0.04 (-63.49%)0.04 (-55.40%)0.01 (-45.59%)209.60 (+124.17%)182.54 (+145.61%)200.30 (+174.01%)140.90 (+125.80%)31.51 (+171.60%)
1531781 — 2026-01-23 18:49:240.13 (n/a)0.11 (n/a)0.11 (n/a)0.09 (n/a)0.02 (n/a)93.50 (n/a)74.32 (n/a)73.10 (n/a)62.40 (n/a)11.60 (n/a)

rms_norm_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.07 (-21.74%)0.05 (-42.82%)0.04 (-53.28%)0.03 (-50.55%)0.02 (+26.37%)264.40 (+102.30%)181.68 (+86.38%)193.10 (+114.08%)109.90 (+27.79%)58.22 (+211.92%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.09 (n/a)0.09 (n/a)0.06 (n/a)0.01 (n/a)130.70 (n/a)97.48 (n/a)90.20 (n/a)86.00 (n/a)18.66 (n/a)

rms_norm_4_cols_2_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.07 (-41.74%)0.04 (-57.50%)0.04 (-66.10%)0.03 (-53.73%)0.01 (-32.68%)271.30 (+116.18%)202.66 (+141.55%)216.70 (+194.83%)125.20 (+71.74%)56.41 (+142.51%)
1531781 — 2026-01-23 18:49:240.11 (n/a)0.10 (n/a)0.11 (n/a)0.07 (n/a)0.02 (n/a)125.50 (n/a)83.90 (n/a)73.50 (n/a)72.90 (n/a)23.26 (n/a)

rms_norm_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.05 (-38.36%)0.04 (-44.56%)0.04 (-49.01%)0.03 (-37.34%)0.01 (-41.41%)246.30 (+59.62%)191.42 (+79.47%)186.00 (+96.20%)152.60 (+62.34%)39.46 (+48.14%)
1531781 — 2026-01-23 18:49:240.09 (n/a)0.08 (n/a)0.09 (n/a)0.05 (n/a)0.01 (n/a)154.30 (n/a)106.66 (n/a)94.80 (n/a)94.00 (n/a)26.64 (n/a)

rms_norm_8_cols_2_channels_2048_tile_128

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.04 (-56.12%)0.04 (-55.06%)0.04 (-59.09%)0.03 (-39.16%)0.00 (-78.66%)275.60 (+64.34%)232.08 (+112.22%)223.90 (+144.43%)202.70 (+128.01%)27.01 (-19.42%)
1531781 — 2026-01-23 18:49:240.09 (n/a)0.08 (n/a)0.09 (n/a)0.05 (n/a)0.02 (n/a)167.70 (n/a)109.36 (n/a)91.60 (n/a)88.90 (n/a)33.52 (n/a)

rope_1_cols_2_channels_4096_tile_4096_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1b2d230 — 2026-01-13 22:38:030.19 (-21.66%)0.17 (-8.53%)0.18 (+2.09%)0.14 (-6.71%)0.02 (-38.07%)174.10 (+7.20%)150.06 (+8.11%)139.60 (-2.04%)129.80 (+27.63%)20.36 (-10.65%)
d48746f — 2026-01-12 22:17:100.24 (n/a)0.18 (n/a)0.17 (n/a)0.15 (n/a)0.04 (n/a)162.40 (n/a)138.80 (n/a)142.50 (n/a)101.70 (n/a)22.79 (n/a)

rope_1c_32rows_512cols_32arows_0m

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.72 (-35.11%)0.61 (-38.30%)0.67 (-36.03%)0.41 (-43.72%)0.12 (-17.45%)237.70 (+77.65%)167.64 (+65.10%)145.90 (+56.38%)137.30 (+54.10%)41.73 (+124.39%)
1531781 — 2026-01-23 18:49:241.10 (n/a)0.99 (n/a)1.05 (n/a)0.73 (n/a)0.15 (n/a)133.80 (n/a)101.54 (n/a)93.30 (n/a)89.10 (n/a)18.60 (n/a)

rope_1c_32rows_512cols_8arows_0m

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.58 (+3.86%)0.44 (-16.84%)0.45 (-14.69%)0.31 (-36.29%)0.11 (+263.29%)241.30 (+56.99%)178.26 (+26.57%)164.30 (+17.27%)127.60 (-3.70%)46.78 (+456.89%)
1531781 — 2026-01-23 18:49:240.56 (n/a)0.52 (n/a)0.53 (n/a)0.48 (n/a)0.03 (n/a)153.70 (n/a)140.84 (n/a)140.10 (n/a)132.50 (n/a)8.40 (n/a)

rope_2_cols_2_channels_4096_tile_2048_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1b2d230 — 2026-01-13 22:38:030.21 (-0.56%)0.17 (-10.62%)0.16 (-13.58%)0.13 (-13.33%)0.03 (+17.05%)183.70 (+15.39%)151.76 (+12.77%)155.20 (+15.73%)117.30 (+0.60%)24.24 (+35.73%)
d48746f — 2026-01-12 22:17:100.21 (n/a)0.19 (n/a)0.18 (n/a)0.15 (n/a)0.02 (n/a)159.20 (n/a)134.58 (n/a)134.10 (n/a)116.60 (n/a)17.86 (n/a)

rope_2c_32rows_512cols_32arows_0m

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.86 (-31.51%)0.69 (-28.52%)0.68 (-25.40%)0.55 (-12.46%)0.11 (-55.85%)179.30 (+14.20%)145.34 (+34.45%)145.20 (+34.07%)114.90 (+46.00%)23.12 (-25.64%)
1531781 — 2026-01-23 18:49:241.25 (n/a)0.97 (n/a)0.91 (n/a)0.63 (n/a)0.25 (n/a)157.00 (n/a)108.10 (n/a)108.30 (n/a)78.70 (n/a)31.09 (n/a)

rope_2c_32rows_512cols_8arows_0m

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.51 (-19.44%)0.42 (-26.19%)0.48 (-14.70%)0.27 (-44.60%)0.10 (+99.38%)271.80 (+80.48%)186.34 (+42.70%)152.80 (+17.18%)145.00 (+24.14%)54.61 (+334.60%)
1531781 — 2026-01-23 18:49:240.63 (n/a)0.57 (n/a)0.57 (n/a)0.49 (n/a)0.05 (n/a)150.60 (n/a)130.58 (n/a)130.40 (n/a)116.80 (n/a)12.57 (n/a)

rope_4_cols_2_channels_4096_tile_1024_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1b2d230 — 2026-01-13 22:38:030.22 (+18.92%)0.17 (+6.24%)0.17 (+5.57%)0.14 (-0.50%)0.03 (+90.77%)171.10 (+0.47%)147.40 (-4.39%)148.10 (-5.31%)110.10 (-15.89%)23.59 (+60.67%)
d48746f — 2026-01-12 22:17:100.19 (n/a)0.16 (n/a)0.16 (n/a)0.14 (n/a)0.02 (n/a)170.30 (n/a)154.16 (n/a)156.40 (n/a)130.90 (n/a)14.69 (n/a)

rope_8_cols_2_channels_4096_tile_512_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1b2d230 — 2026-01-13 22:38:030.17 (-1.30%)0.15 (+6.63%)0.15 (-8.78%)0.14 (+39.68%)0.01 (-67.35%)174.90 (-28.41%)163.42 (-10.36%)167.40 (+9.63%)146.50 (+1.31%)11.19 (-76.17%)
d48746f — 2026-01-12 22:17:100.17 (n/a)0.14 (n/a)0.16 (n/a)0.10 (n/a)0.03 (n/a)244.30 (n/a)182.30 (n/a)152.70 (n/a)144.60 (n/a)46.98 (n/a)

rope_8c_32rows_512cols_32arows_0m

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.54 (-59.40%)0.47 (-47.64%)0.43 (-49.43%)0.42 (-30.23%)0.06 (-78.28%)234.40 (+43.36%)212.48 (+81.02%)228.40 (+97.75%)181.20 (+146.20%)25.41 (-20.79%)
1531781 — 2026-01-23 18:49:241.34 (n/a)0.89 (n/a)0.85 (n/a)0.60 (n/a)0.27 (n/a)163.50 (n/a)117.38 (n/a)115.50 (n/a)73.60 (n/a)32.08 (n/a)

rope_8c_32rows_512cols_8arows_0m

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.59 (+7.84%)0.47 (-8.13%)0.48 (-8.27%)0.28 (-41.33%)0.13 (+377.80%)260.60 (+70.44%)167.26 (+16.70%)152.30 (+9.02%)125.60 (-7.31%)55.69 (+645.99%)
1531781 — 2026-01-23 18:49:240.54 (n/a)0.52 (n/a)0.53 (n/a)0.48 (n/a)0.03 (n/a)152.90 (n/a)143.32 (n/a)139.70 (n/a)135.50 (n/a)7.46 (n/a)

sigmoid_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.06 (-48.91%)0.05 (-44.76%)0.06 (-52.94%)0.05 (-8.71%)0.01 (-81.59%)167.80 (+9.53%)150.16 (+65.16%)145.70 (+112.39%)133.80 (+95.90%)15.16 (-58.82%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.10 (n/a)0.12 (n/a)0.05 (n/a)0.03 (n/a)153.20 (n/a)90.92 (n/a)68.60 (n/a)68.30 (n/a)36.80 (n/a)

sigmoid_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.08 (-37.47%)0.06 (-39.97%)0.06 (-45.85%)0.04 (-33.55%)0.01 (-49.56%)212.00 (+50.57%)145.46 (+61.91%)127.30 (+84.76%)109.00 (+60.06%)41.38 (+29.15%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.10 (n/a)0.12 (n/a)0.06 (n/a)0.03 (n/a)140.80 (n/a)89.84 (n/a)68.90 (n/a)68.10 (n/a)32.04 (n/a)

sigmoid_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.07 (-24.39%)0.06 (-27.99%)0.05 (-43.41%)0.05 (+5.28%)0.01 (-51.48%)160.90 (-5.02%)140.96 (+32.26%)154.50 (+76.77%)111.10 (+32.26%)22.26 (-38.80%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.08 (n/a)0.09 (n/a)0.05 (n/a)0.02 (n/a)169.40 (n/a)106.58 (n/a)87.40 (n/a)84.00 (n/a)36.37 (n/a)

sigmoid_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.06 (-50.16%)0.05 (-49.18%)0.05 (-58.11%)0.04 (-22.01%)0.01 (-73.57%)201.70 (+28.23%)161.88 (+80.39%)164.00 (+138.72%)132.30 (+100.76%)26.28 (-32.36%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.10 (n/a)0.12 (n/a)0.05 (n/a)0.03 (n/a)157.30 (n/a)89.74 (n/a)68.70 (n/a)65.90 (n/a)38.85 (n/a)

silu_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.09 (-15.50%)0.06 (-30.36%)0.06 (-43.17%)0.04 (-38.66%)0.02 (-20.68%)233.70 (+63.08%)152.52 (+45.42%)144.70 (+76.03%)95.40 (+18.36%)50.67 (+57.19%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.08 (n/a)0.10 (n/a)0.06 (n/a)0.02 (n/a)143.30 (n/a)104.88 (n/a)82.20 (n/a)80.60 (n/a)32.24 (n/a)

silu_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.07 (-34.84%)0.05 (-38.23%)0.06 (-43.42%)0.03 (-45.75%)0.01 (-40.05%)256.10 (+84.38%)167.56 (+62.52%)147.10 (+76.80%)125.20 (+53.43%)51.49 (+81.91%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.08 (n/a)0.10 (n/a)0.06 (n/a)0.02 (n/a)138.90 (n/a)103.10 (n/a)83.20 (n/a)81.60 (n/a)28.30 (n/a)

silu_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.05 (-55.30%)0.05 (-54.53%)0.05 (-59.52%)0.04 (-27.26%)0.00 (-84.33%)188.80 (+37.41%)171.18 (+105.35%)172.90 (+147.00%)152.30 (+123.64%)14.28 (-52.81%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.11 (n/a)0.12 (n/a)0.06 (n/a)0.03 (n/a)137.40 (n/a)83.36 (n/a)70.00 (n/a)68.10 (n/a)30.27 (n/a)

silu_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.08 (-28.66%)0.05 (-43.00%)0.05 (-48.77%)0.04 (-43.77%)0.02 (+6.92%)195.30 (+77.87%)162.00 (+81.61%)171.70 (+95.34%)99.70 (+40.23%)36.31 (+150.31%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.09 (n/a)0.09 (n/a)0.07 (n/a)0.02 (n/a)109.80 (n/a)89.20 (n/a)87.90 (n/a)71.10 (n/a)14.50 (n/a)

softmax_1_cols_2_channels_4096_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.21 (+20.19%)0.16 (+13.84%)0.16 (+8.19%)0.12 (+26.61%)0.04 (+17.18%)135.50 (-20.99%)104.68 (-12.65%)102.70 (-7.56%)76.50 (-16.85%)22.55 (-26.21%)
0f0da0b — 2026-01-14 16:31:000.18 (n/a)0.14 (n/a)0.15 (n/a)0.10 (n/a)0.03 (n/a)171.50 (n/a)119.84 (n/a)111.10 (n/a)92.00 (n/a)30.56 (n/a)

softmax_2_cols_2_channels_32768_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:371.05 (n/a)0.93 (n/a)0.97 (n/a)0.68 (n/a)0.15 (n/a)193.60 (n/a)143.86 (n/a)134.90 (n/a)124.50 (n/a)28.21 (n/a)

softmax_2_cols_2_channels_32768_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.91 (n/a)0.84 (n/a)0.87 (n/a)0.68 (n/a)0.09 (n/a)192.40 (n/a)158.20 (n/a)150.70 (n/a)144.20 (n/a)19.41 (n/a)

softmax_2_cols_2_channels_4096_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.22 (+40.26%)0.18 (+43.33%)0.20 (+64.08%)0.13 (+24.34%)0.04 (+68.98%)127.80 (-19.62%)94.74 (-28.95%)83.40 (-39.08%)73.90 (-28.67%)24.46 (-5.31%)
0f0da0b — 2026-01-14 16:31:000.16 (n/a)0.13 (n/a)0.12 (n/a)0.10 (n/a)0.03 (n/a)159.00 (n/a)133.34 (n/a)136.90 (n/a)103.60 (n/a)25.83 (n/a)

softmax_2_cols_2_channels_4096_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.24 (+85.21%)0.17 (+69.97%)0.15 (+54.16%)0.11 (+37.21%)0.05 (+203.82%)148.10 (-27.12%)102.78 (-37.88%)106.70 (-35.10%)69.60 (-45.96%)31.81 (+15.98%)
0f0da0b — 2026-01-14 16:31:000.13 (n/a)0.10 (n/a)0.10 (n/a)0.08 (n/a)0.02 (n/a)203.20 (n/a)165.46 (n/a)164.40 (n/a)128.80 (n/a)27.43 (n/a)

softmax_4_cols_4_channels_32768_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.94 (n/a)0.79 (n/a)0.76 (n/a)0.66 (n/a)0.12 (n/a)197.60 (n/a)168.34 (n/a)173.10 (n/a)139.20 (n/a)25.40 (n/a)

swiglu

No metrics available.

swiglu_decode_1x2048x2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.00 (-15.46%)0.00 (-27.20%)0.00 (-28.17%)0.00 (-33.90%)0.00 (+1834.83%)5187.20 (+51.28%)4720.84 (+38.38%)4750.90 (+39.22%)4015.10 (+18.29%)390.23 (+3298.45%)
0a6c11c — 2025-12-04 00:39:100.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)3428.80 (n/a)3411.62 (n/a)3412.60 (n/a)3394.20 (n/a)11.48 (n/a)

swiglu_decode_1x2048x2048_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:37n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)
1531781 — 2026-01-23 18:49:240.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)5111.40 (n/a)4324.88 (n/a)4736.10 (n/a)3410.70 (n/a)843.97 (n/a)

swiglu_prefill_256x2048x2048_0

No metrics available.

tanh_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.07 (-23.79%)0.06 (-27.99%)0.05 (-26.22%)0.04 (-35.23%)0.01 (-15.50%)217.40 (+54.40%)154.24 (+41.32%)157.10 (+35.55%)112.30 (+31.19%)41.64 (+77.08%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.08 (n/a)0.07 (n/a)0.06 (n/a)0.02 (n/a)140.80 (n/a)109.14 (n/a)115.90 (n/a)85.60 (n/a)23.52 (n/a)

tanh_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.07 (-37.14%)0.05 (-47.30%)0.04 (-53.69%)0.04 (-54.28%)0.01 (+31.49%)226.00 (+118.78%)174.26 (+97.98%)189.50 (+115.83%)123.70 (+59.00%)42.65 (+342.59%)
1531781 — 2026-01-23 18:49:240.11 (n/a)0.09 (n/a)0.09 (n/a)0.08 (n/a)0.01 (n/a)103.30 (n/a)88.02 (n/a)87.80 (n/a)77.80 (n/a)9.64 (n/a)

tanh_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.07 (-30.81%)0.06 (-37.36%)0.06 (-41.98%)0.04 (-33.46%)0.01 (-33.77%)182.40 (+50.25%)148.96 (+59.49%)148.80 (+72.42%)122.30 (+44.56%)22.55 (+42.66%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.09 (n/a)0.09 (n/a)0.07 (n/a)0.01 (n/a)121.40 (n/a)93.40 (n/a)86.30 (n/a)84.60 (n/a)15.80 (n/a)

tanh_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.05 (-57.42%)0.05 (-48.61%)0.05 (-54.13%)0.04 (-17.83%)0.00 (-88.11%)199.30 (+21.67%)181.22 (+76.11%)175.20 (+117.91%)168.30 (+134.73%)13.88 (-65.30%)
1531781 — 2026-01-23 18:49:240.11 (n/a)0.09 (n/a)0.10 (n/a)0.05 (n/a)0.03 (n/a)163.80 (n/a)102.90 (n/a)80.40 (n/a)71.70 (n/a)39.99 (n/a)

transpose_2048_M_64_N_1_cols_1_channels_64_m_64_n_8_s

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:493.93 (-7.02%)3.68 (+2.48%)3.66 (+1.82%)3.45 (+15.62%)0.15 (-63.64%)151.80 (-13.50%)142.70 (-3.59%)143.40 (-1.78%)133.50 (+7.49%)5.90 (-66.22%)
0a6c11c — 2025-12-04 00:39:104.22 (n/a)3.59 (n/a)3.59 (n/a)2.99 (n/a)0.42 (n/a)175.50 (n/a)148.02 (n/a)146.00 (n/a)124.20 (n/a)17.46 (n/a)

transpose_2048_M_64_N_1_cols_1_channels_64_m_64_n_8_s0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:373.84 (-22.42%)3.16 (-27.20%)3.00 (-31.41%)2.52 (-32.65%)0.52 (+5.02%)208.40 (+48.54%)169.48 (+38.94%)174.80 (+45.79%)136.60 (+28.87%)28.15 (+98.86%)
1531781 — 2026-01-23 18:49:244.95 (n/a)4.34 (n/a)4.37 (n/a)3.74 (n/a)0.50 (n/a)140.30 (n/a)121.98 (n/a)119.90 (n/a)106.00 (n/a)14.15 (n/a)

transpose_2048_M_64_N_1_cols_2_channels_64_m_64_n_8_s

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:495.11 (+15.57%)3.85 (-3.32%)3.40 (-18.64%)3.29 (+7.09%)0.71 (+49.14%)159.60 (-6.61%)140.18 (+4.80%)154.20 (+22.97%)102.60 (-13.49%)22.70 (+19.77%)
0a6c11c — 2025-12-04 00:39:104.42 (n/a)3.99 (n/a)4.18 (n/a)3.07 (n/a)0.48 (n/a)170.90 (n/a)133.76 (n/a)125.40 (n/a)118.60 (n/a)18.95 (n/a)

transpose_2048_M_64_N_1_cols_2_channels_64_m_64_n_8_s0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:373.95 (-35.00%)3.05 (-38.83%)2.89 (-43.13%)2.68 (-34.66%)0.52 (-29.82%)196.00 (+53.01%)175.16 (+63.92%)181.20 (+75.92%)132.80 (+53.88%)25.52 (+62.87%)
1531781 — 2026-01-23 18:49:246.07 (n/a)4.99 (n/a)5.09 (n/a)4.09 (n/a)0.74 (n/a)128.10 (n/a)106.86 (n/a)103.00 (n/a)86.30 (n/a)15.67 (n/a)

weighted_rms_norm_1_cols_2_channels_2048_weights_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.10 (-7.65%)0.08 (-9.46%)0.08 (-12.45%)0.05 (+3.86%)0.02 (-22.90%)223.90 (+44.45%)161.74 (+61.87%)153.50 (+71.32%)124.20 (+62.35%)37.03 (+18.09%)
1531781 — 2026-01-23 18:49:240.11 (n/a)0.09 (n/a)0.09 (n/a)0.05 (n/a)0.02 (n/a)155.00 (n/a)99.92 (n/a)89.60 (n/a)76.50 (n/a)31.36 (n/a)

weighted_rms_norm_2_cols_2_channels_2048_weights_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.09 (-9.80%)0.06 (-26.47%)0.07 (-29.62%)0.04 (-10.93%)0.02 (-15.84%)233.90 (+40.31%)172.66 (+68.09%)157.10 (+77.51%)115.20 (+38.63%)45.63 (+27.21%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.09 (n/a)0.09 (n/a)0.05 (n/a)0.02 (n/a)166.70 (n/a)102.72 (n/a)88.50 (n/a)83.10 (n/a)35.87 (n/a)

weighted_rms_norm_4_cols_2_channels_2048_weights_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.06 (-32.71%)0.05 (-40.52%)0.05 (-42.55%)0.04 (-39.87%)0.01 (-22.49%)222.70 (+87.14%)183.56 (+90.26%)176.00 (+95.77%)149.30 (+67.19%)27.63 (+115.77%)
1531781 — 2026-01-23 18:49:240.09 (n/a)0.09 (n/a)0.09 (n/a)0.07 (n/a)0.01 (n/a)119.00 (n/a)96.48 (n/a)89.90 (n/a)89.30 (n/a)12.81 (n/a)

weighted_rms_norm_8_cols_2_channels_2048_weights_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
4a70412 — 2026-02-09 15:12:370.05 (-55.65%)0.04 (-54.81%)0.04 (-59.75%)0.04 (-32.08%)0.00 (-83.00%)212.40 (+56.41%)202.26 (+124.93%)211.20 (+164.00%)178.10 (+139.70%)14.86 (-42.33%)
1531781 — 2026-01-23 18:49:240.11 (n/a)0.10 (n/a)0.10 (n/a)0.06 (n/a)0.02 (n/a)135.80 (n/a)89.92 (n/a)80.00 (n/a)74.30 (n/a)25.76 (n/a)

@github-actions
Copy link
Contributor

github-actions bot commented Feb 9, 2026

📊 Test Results for Small Benchmark/Test Suite

71486ae (2026_02_09_15_38_10)

IRONCLAD

Tested on 2026_02_09_15_38_10 at commit 71486ae.

Test Checks Latency (mean)Bandwidth (mean)GFLOP/s (mean)
axpy_1_cols_2_channels_2048_tile_2048_3.0_0 ✅ 5/5 161.38 0.08 n/a
axpy_2_cols_2_channels_2048_tile_1024_3.0_0 ✅ 5/5 193.40 0.07 n/a
axpy_4_cols_2_channels_2048_tile_512_3.0_0 ✅ 5/5 176.10 0.07 n/a
axpy_8_cols_2_channels_2048_tile_256_3.0_0 ✅ 5/5 195.34 0.06 n/a
dequant_1_cols_1_channels_2048_tile_2048_0 ✅ 5/5 129.88 0.04 n/a
dequant_1_cols_2_channels_2048_tile_1024_0 ✅ 5/5 162.46 0.03 n/a
dequant_2_cols_1_channels_2048_tile_1024_0 ✅ 5/5 145.86 0.04 n/a
dequant_2_cols_2_channels_2048_tile_512_0 ✅ 5/5 143.40 0.04 n/a
dequant_4_cols_1_channels_2048_tile_512_0 ✅ 5/5 136.64 0.04 n/a
dequant_4_cols_2_channels_2048_tile_256_0 ✅ 5/5 169.94 0.03 n/a
dequant_8_cols_1_channels_2048_tile_256_0 ✅ 5/5 167.36 0.03 n/a
dequant_8_cols_2_channels_2048_tile_128_0 ✅ 5/5 202.86 0.03 n/a
eltwise_add_1_cols_2_channels_2048_tile_2048 ✅ 5/5 122.52 0.10 n/a
eltwise_add_2_cols_2_channels_2048_tile_1024 ✅ 5/5 143.40 0.09 n/a
eltwise_add_4_cols_2_channels_2048_tile_512 ✅ 5/5 157.30 0.08 n/a
eltwise_add_8_cols_2_channels_2048_tile_256 ✅ 5/5 177.62 0.07 n/a
eltwise_mul_1_cols_2_channels_2048_tile_2048 ✅ 5/5 130.12 0.10 n/a
eltwise_mul_2_cols_2_channels_2048_tile_1024 ✅ 5/5 131.84 0.10 n/a
eltwise_mul_4_cols_2_channels_2048_tile_512 ✅ 5/5 146.96 0.09 n/a
eltwise_mul_8_cols_2_channels_2048_tile_256 ✅ 5/5 209.88 0.06 n/a
gelu_1_cols_1_channels_2048_tile_2048 ✅ 5/5 149.06 0.06 n/a
gelu_1_cols_2_channels_2048_tile_1024 ✅ 5/5 154.00 0.06 n/a
gelu_2_cols_1_channels_2048_tile_1024 ✅ 5/5 188.08 0.04 n/a
gelu_2_cols_2_channels_2048_tile_512 ✅ 5/5 182.66 0.05 n/a
gelu_4_cols_1_channels_2048_tile_512 ✅ 5/5 165.92 0.05 n/a
gelu_4_cols_2_channels_2048_tile_256 ✅ 5/5 162.56 0.05 n/a
gelu_8_cols_1_channels_2048_tile_256 ✅ 5/5 213.04 0.04 n/a
gelu_8_cols_2_channels_2048_tile_128 ✅ 5/5 208.28 0.04 n/a
gemm_1792x896x1152_64x32x48_8cols_ccolmaj ✅ 5/5 2160.70 4.37 n/a
gemm_192x384x64_48x96x16_4cols ✅ 5/5 218.34 1.02 n/a
gemm_192x384x64_48x96x16_4cols_bcolmaj_ccolmaj ✅ 5/5 171.08 1.38 n/a
gemm_2048x2048x2048_64x64x64_1cols ✅ 5/5 48561.24 0.52 n/a
gemm_2048x2048x2048_64x64x64_2cols_bcolmaj ✅ 5/5 27688.98 0.91 n/a
gemm_2048x2048x2048_64x64x64_8cols_bcolmaj_ccolmaj ✅ 5/5 7698.64 3.27 n/a
gemm_384x1536x1792_32x48x64_4cols_bcolmaj ✅ 5/5 2101.42 3.85 n/a
gemm_896x1792x640_32x64x80_8cols_ccolmaj ✅ 5/5 1432.10 4.65 n/a
layer_norm_1_cols_1_channels_2048_tile_2048 ✅ 5/5 141.28 0.06 n/a
layer_norm_1_cols_2_channels_2048_tile_1024 ✅ 5/5 171.50 0.05 n/a
layer_norm_2_cols_1_channels_2048_tile_1024 ✅ 5/5 169.32 0.05 n/a
layer_norm_2_cols_2_channels_2048_tile_512 ✅ 5/5 146.58 0.06 n/a
layer_norm_4_cols_1_channels_2048_tile_512 ✅ 5/5 166.18 0.05 n/a
layer_norm_4_cols_2_channels_2048_tile_256 ✅ 5/5 171.66 0.05 n/a
layer_norm_8_cols_1_channels_2048_tile_256 ✅ 5/5 170.74 0.05 n/a
layer_norm_8_cols_2_channels_2048_tile_128 ✅ 5/5 157.08 0.05 n/a
matrix_vector_mul_128x128_32tsi_128tso_1col0 ✅ 5/5 n/a 0.23 n/a
matrix_vector_mul_2048x8192_1tsi_1024tso_2col0 ✅ 5/5 n/a 24.75 n/a
matrix_vector_mul_2048x8192_1tsi_2048tso_1col0 ✅ 5/5 n/a 12.93 n/a
matrix_vector_mul_2048x8192_1tsi_256tso_8col0 ✅ 5/5 n/a 44.57 n/a
matrix_vector_mul_2048x8192_1tsi_512tso_4col0 ✅ 5/5 n/a 40.93 n/a
matrix_vector_mul_8192x2048_4tsi_1024tso_1col0 ✅ 5/5 n/a 13.52 n/a
matrix_vector_mul_8192x2048_4tsi_1024tso_2col0 ✅ 5/5 n/a 24.48 n/a
matrix_vector_mul_8192x2048_4tsi_1024tso_4col0 ✅ 5/5 n/a 40.94 n/a
matrix_vector_mul_8192x2048_4tsi_1024tso_8col0 ✅ 5/5 n/a 44.18 n/a
mem_copy_16_cores_2_chans_2048_tile_128_False0 ✅ 5/5 205.32 0.04 n/a
mem_copy_1_cores_1_chans_2048_tile_2048_False0 ✅ 5/5 176.06 0.05 n/a
mem_copy_2_cores_1_chans_2048_tile_1024_False0 ✅ 5/5 181.16 0.05 n/a
mem_copy_2_cores_2_chans_2048_tile_1024_False0 ✅ 5/5 158.18 0.05 n/a
mem_copy_4_cores_1_chans_2048_tile_512_False0 ✅ 5/5 153.12 0.06 n/a
mem_copy_4_cores_2_chans_2048_tile_512_False0 ✅ 5/5 153.98 0.06 n/a
mem_copy_8_cores_1_chans_2048_tile_256_False0 ✅ 5/5 160.22 0.05 n/a
mem_copy_8_cores_2_chans_2048_tile_256_False0 ✅ 5/5 168.72 0.05 n/a
mha0 ✅ 5/5 47512.04 0.18 n/a
relu_1_cols_1_channels_2048_tile_2048 ✅ 5/5 127.84 0.07 n/a
relu_2_cols_1_channels_2048_tile_1024 ✅ 5/5 112.34 0.08 n/a
relu_4_cols_1_channels_2048_tile_512 ✅ 5/5 130.66 0.06 n/a
relu_8_cols_1_channels_2048_tile_256 ✅ 5/5 143.02 0.06 n/a
rms_norm_1_cols_1_channels_2048_tile_2048 ✅ 5/5 139.62 0.06 n/a
rms_norm_1_cols_2_channels_2048_tile_1024 ✅ 5/5 163.12 0.05 n/a
rms_norm_2_cols_1_channels_2048_tile_1024 ✅ 5/5 159.94 0.05 n/a
rms_norm_2_cols_2_channels_2048_tile_512 ✅ 5/5 174.28 0.05 n/a
rms_norm_4_cols_1_channels_2048_tile_512 ✅ 5/5 148.74 0.06 n/a
rms_norm_4_cols_2_channels_2048_tile_256 ✅ 5/5 171.76 0.05 n/a
rms_norm_8_cols_1_channels_2048_tile_256 ✅ 5/5 168.32 0.05 n/a
rms_norm_8_cols_2_channels_2048_tile_128 ✅ 5/5 273.56 0.03 n/a
rope_1c_32rows_512cols_32arows_0m ✅ 5/5 169.58 0.61 n/a
rope_1c_32rows_512cols_8arows_0m ✅ 5/5 174.04 0.45 n/a
rope_2c_32rows_512cols_32arows_0m ✅ 5/5 161.90 0.64 n/a
rope_2c_32rows_512cols_8arows_0m ✅ 5/5 185.16 0.43 n/a
rope_8c_32rows_512cols_32arows_0m ✅ 5/5 187.96 0.53 n/a
rope_8c_32rows_512cols_8arows_0m ✅ 5/5 159.22 0.50 n/a
sigmoid_1_cols_1_channels_2048_tile_2048 ✅ 5/5 135.80 0.06 n/a
sigmoid_2_cols_1_channels_2048_tile_1024 ✅ 5/5 148.78 0.06 n/a
sigmoid_4_cols_1_channels_2048_tile_512 ✅ 5/5 138.60 0.06 n/a
sigmoid_8_cols_1_channels_2048_tile_256 ✅ 5/5 183.72 0.04 n/a
silu_1_cols_1_channels_2048_tile_2048 ✅ 5/5 141.66 0.06 n/a
silu_2_cols_1_channels_2048_tile_1024 ✅ 5/5 162.14 0.05 n/a
silu_4_cols_1_channels_2048_tile_512 ✅ 5/5 159.84 0.05 n/a
silu_8_cols_1_channels_2048_tile_256 ✅ 5/5 210.24 0.04 n/a
softmax_2_cols_2_channels_32768_tile_1024 ✅ 5/5 161.36 0.88 n/a
softmax_2_cols_2_channels_32768_tile_512 ✅ 5/5 170.88 0.79 n/a
softmax_4_cols_4_channels_32768_tile_2048 ✅ 5/5 289.12 0.50 n/a
swiglu_decode_1x2048x2048_0 ✅ 5/5 n/a n/a n/a
swiglu_prefill_256x2048x2048_0 ❌ 0/5 n/a n/a n/a
tanh_1_cols_1_channels_2048_tile_2048 ✅ 5/5 133.28 0.06 n/a
tanh_2_cols_1_channels_2048_tile_1024 ✅ 5/5 152.60 0.06 n/a
tanh_4_cols_1_channels_2048_tile_512 ✅ 5/5 151.08 0.06 n/a
tanh_8_cols_1_channels_2048_tile_256 ✅ 5/5 171.38 0.05 n/a
transpose_2048_M_64_N_1_cols_1_channels_64_m_64_n_8_s0 ✅ 5/5 156.88 3.36 n/a
transpose_2048_M_64_N_1_cols_2_channels_64_m_64_n_8_s0 ✅ 5/5 157.88 3.44 n/a
weighted_rms_norm_1_cols_2_channels_2048_weights_2048 ✅ 5/5 171.38 0.07 n/a
weighted_rms_norm_2_cols_2_channels_2048_weights_1024 ✅ 5/5 152.36 0.07 n/a
weighted_rms_norm_4_cols_2_channels_2048_weights_512 ✅ 5/5 180.20 0.05 n/a
weighted_rms_norm_8_cols_2_channels_2048_weights_256 ✅ 5/5 180.86 0.05 n/a
📈 Trends (vs main branch) for Small Benchmark/Test Suite

71486ae (2026_02_09_15_38_10)

IRONCLAD Trends

axpy_1_cols_2_channels_2048_tile_2048_3.0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.13 (-1.96%)0.11 (+2.52%)0.10 (+1.86%)0.08 (-1.91%)0.02 (-1.72%)162.40 (+1.95%)120.96 (-2.36%)117.00 (-1.85%)94.40 (+2.05%)23.11 (+4.09%)
0a6c11c — 2025-12-04 00:39:100.13 (n/a)0.10 (n/a)0.10 (n/a)0.08 (n/a)0.02 (n/a)159.30 (n/a)123.88 (n/a)119.20 (n/a)92.50 (n/a)22.21 (n/a)

axpy_1_cols_2_channels_2048_tile_2048_3.0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.10 (-37.45%)0.08 (-37.51%)0.08 (-27.28%)0.05 (-47.05%)0.02 (-39.15%)224.40 (+88.89%)161.38 (+60.90%)151.90 (+37.59%)124.70 (+59.87%)39.10 (+90.84%)
1531781 — 2026-01-23 18:49:240.16 (n/a)0.13 (n/a)0.11 (n/a)0.10 (n/a)0.03 (n/a)118.80 (n/a)100.30 (n/a)110.40 (n/a)78.00 (n/a)20.49 (n/a)

axpy_2_cols_2_channels_2048_tile_1024_3.0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.11 (-26.77%)0.09 (-9.09%)0.09 (-10.81%)0.08 (+8.25%)0.01 (-55.42%)152.00 (-7.60%)135.46 (+6.28%)142.80 (+12.18%)115.00 (+36.58%)15.03 (-41.36%)
0a6c11c — 2025-12-04 00:39:100.15 (n/a)0.10 (n/a)0.10 (n/a)0.07 (n/a)0.02 (n/a)164.50 (n/a)127.46 (n/a)127.30 (n/a)84.20 (n/a)25.64 (n/a)

axpy_2_cols_2_channels_2048_tile_1024_3.0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.08 (-40.41%)0.07 (-46.92%)0.07 (-51.71%)0.05 (-35.04%)0.01 (-49.94%)241.50 (+53.92%)193.40 (+85.14%)186.90 (+106.98%)147.60 (+67.92%)37.56 (+27.25%)
1531781 — 2026-01-23 18:49:240.14 (n/a)0.12 (n/a)0.14 (n/a)0.08 (n/a)0.03 (n/a)156.90 (n/a)104.46 (n/a)90.30 (n/a)87.90 (n/a)29.52 (n/a)

axpy_4_cols_2_channels_2048_tile_512_3.0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.12 (-10.21%)0.10 (-8.41%)0.10 (-6.20%)0.07 (+3.52%)0.02 (-23.48%)166.70 (-3.36%)128.78 (+7.33%)121.80 (+6.65%)103.90 (+11.36%)22.95 (-18.82%)
0a6c11c — 2025-12-04 00:39:100.13 (n/a)0.11 (n/a)0.11 (n/a)0.07 (n/a)0.02 (n/a)172.50 (n/a)119.98 (n/a)114.20 (n/a)93.30 (n/a)28.27 (n/a)

axpy_4_cols_2_channels_2048_tile_512_3.0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.10 (-37.31%)0.07 (-47.22%)0.07 (-55.15%)0.06 (-37.17%)0.02 (-32.35%)213.40 (+59.13%)176.10 (+90.50%)186.00 (+123.02%)125.40 (+59.54%)39.57 (+69.24%)
1531781 — 2026-01-23 18:49:240.16 (n/a)0.14 (n/a)0.15 (n/a)0.09 (n/a)0.03 (n/a)134.10 (n/a)92.44 (n/a)83.40 (n/a)78.60 (n/a)23.38 (n/a)

axpy_8_cols_2_channels_2048_tile_256_3.0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.10 (-16.19%)0.08 (-10.40%)0.09 (-2.09%)0.07 (-17.38%)0.01 (-10.61%)184.40 (+21.00%)153.18 (+11.99%)144.10 (+2.20%)127.10 (+19.34%)21.26 (+34.76%)
0a6c11c — 2025-12-04 00:39:100.12 (n/a)0.09 (n/a)0.09 (n/a)0.08 (n/a)0.01 (n/a)152.40 (n/a)136.78 (n/a)141.00 (n/a)106.50 (n/a)15.78 (n/a)

axpy_8_cols_2_channels_2048_tile_256_3.0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.07 (-60.64%)0.06 (-55.71%)0.06 (-58.75%)0.06 (-31.95%)0.00 (-88.67%)215.80 (+47.00%)195.34 (+111.77%)193.80 (+142.25%)183.80 (+154.22%)12.77 (-58.90%)
1531781 — 2026-01-23 18:49:240.17 (n/a)0.14 (n/a)0.15 (n/a)0.08 (n/a)0.03 (n/a)146.80 (n/a)92.24 (n/a)80.00 (n/a)72.30 (n/a)31.07 (n/a)

dequant_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.05 (-18.83%)0.04 (-5.21%)0.04 (-2.69%)0.03 (+37.10%)0.00 (-53.98%)160.40 (-27.06%)141.12 (-0.91%)136.40 (+2.79%)114.30 (+23.17%)17.37 (-59.06%)
0a6c11c — 2025-12-04 00:39:100.06 (n/a)0.04 (n/a)0.04 (n/a)0.02 (n/a)0.01 (n/a)219.90 (n/a)142.42 (n/a)132.70 (n/a)92.80 (n/a)42.43 (n/a)

dequant_1_cols_1_channels_2048_tile_2048_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.05 (-17.68%)0.04 (-11.28%)0.04 (+0.01%)0.03 (-18.28%)0.01 (-16.15%)153.50 (+22.31%)129.88 (+12.86%)121.60 (+0.00%)107.00 (+21.45%)19.97 (+29.71%)
1531781 — 2026-01-23 18:49:240.06 (n/a)0.05 (n/a)0.04 (n/a)0.04 (n/a)0.01 (n/a)125.50 (n/a)115.08 (n/a)121.60 (n/a)88.10 (n/a)15.40 (n/a)

dequant_1_cols_2_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.06 (-13.28%)0.05 (-8.93%)0.06 (-13.59%)0.04 (+4.63%)0.01 (-39.10%)133.60 (-4.43%)106.08 (+5.85%)95.10 (+15.69%)87.90 (+15.35%)17.90 (-31.97%)
0a6c11c — 2025-12-04 00:39:100.07 (n/a)0.06 (n/a)0.06 (n/a)0.04 (n/a)0.01 (n/a)139.80 (n/a)100.22 (n/a)82.20 (n/a)76.20 (n/a)26.31 (n/a)

dequant_1_cols_2_channels_2048_tile_1024_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.05 (-27.24%)0.03 (-28.56%)0.03 (-28.28%)0.02 (-25.27%)0.01 (-20.19%)219.10 (+33.84%)162.46 (+41.69%)155.20 (+39.44%)105.10 (+37.39%)47.10 (+48.92%)
1531781 — 2026-01-23 18:49:240.07 (n/a)0.05 (n/a)0.05 (n/a)0.03 (n/a)0.01 (n/a)163.70 (n/a)114.66 (n/a)111.30 (n/a)76.50 (n/a)31.63 (n/a)

dequant_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.07 (+5.66%)0.05 (+1.21%)0.05 (+0.61%)0.04 (-15.02%)0.01 (+58.42%)143.40 (+17.64%)101.02 (+1.49%)97.60 (-0.61%)78.90 (-5.40%)22.56 (+78.52%)
0a6c11c — 2025-12-04 00:39:100.06 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)121.90 (n/a)99.54 (n/a)98.20 (n/a)83.40 (n/a)12.64 (n/a)

dequant_2_cols_1_channels_2048_tile_1024_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.05 (-1.96%)0.04 (-18.86%)0.04 (-25.82%)0.02 (-32.71%)0.01 (+49.76%)219.60 (+48.58%)145.86 (+28.92%)143.50 (+34.74%)103.50 (+2.07%)44.84 (+128.61%)
1531781 — 2026-01-23 18:49:240.05 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)147.80 (n/a)113.14 (n/a)106.50 (n/a)101.40 (n/a)19.61 (n/a)

dequant_2_cols_2_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.05 (+17.06%)0.04 (+8.98%)0.05 (+21.37%)0.03 (-11.65%)0.01 (+301.89%)159.70 (+13.18%)126.32 (-5.68%)109.00 (-17.67%)106.80 (-14.56%)22.48 (+280.15%)
0a6c11c — 2025-12-04 00:39:100.04 (n/a)0.04 (n/a)0.04 (n/a)0.04 (n/a)0.00 (n/a)141.10 (n/a)133.92 (n/a)132.40 (n/a)125.00 (n/a)5.91 (n/a)

dequant_2_cols_2_channels_2048_tile_512_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.05 (-5.96%)0.04 (-13.47%)0.04 (-27.34%)0.03 (-22.59%)0.01 (+11.57%)204.60 (+29.25%)143.40 (+18.14%)146.20 (+37.66%)104.40 (+6.31%)39.57 (+51.46%)
1531781 — 2026-01-23 18:49:240.05 (n/a)0.04 (n/a)0.05 (n/a)0.03 (n/a)0.01 (n/a)158.30 (n/a)121.38 (n/a)106.20 (n/a)98.20 (n/a)26.13 (n/a)

dequant_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.06 (+20.24%)0.05 (-3.19%)0.05 (-6.73%)0.03 (-20.69%)0.01 (+194.26%)156.10 (+26.09%)120.86 (+7.15%)114.90 (+7.18%)88.30 (-16.85%)24.40 (+214.82%)
0a6c11c — 2025-12-04 00:39:100.05 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.00 (n/a)123.80 (n/a)112.80 (n/a)107.20 (n/a)106.20 (n/a)7.75 (n/a)

dequant_4_cols_1_channels_2048_tile_512_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.05 (-23.66%)0.04 (-19.90%)0.04 (-23.23%)0.03 (-1.84%)0.01 (-49.27%)167.80 (+1.82%)136.64 (+20.26%)134.50 (+30.33%)111.10 (+31.01%)22.03 (-32.32%)
1531781 — 2026-01-23 18:49:240.06 (n/a)0.05 (n/a)0.05 (n/a)0.03 (n/a)0.01 (n/a)164.80 (n/a)113.62 (n/a)103.20 (n/a)84.80 (n/a)32.55 (n/a)

dequant_4_cols_2_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.05 (-6.09%)0.04 (-8.04%)0.04 (-12.46%)0.03 (-0.41%)0.01 (-20.19%)168.10 (+0.42%)135.74 (+7.44%)132.20 (+14.26%)107.40 (+6.44%)22.91 (-12.67%)
0a6c11c — 2025-12-04 00:39:100.05 (n/a)0.04 (n/a)0.05 (n/a)0.03 (n/a)0.01 (n/a)167.40 (n/a)126.34 (n/a)115.70 (n/a)100.90 (n/a)26.23 (n/a)

dequant_4_cols_2_channels_2048_tile_256_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.05 (-27.37%)0.03 (-36.18%)0.03 (-50.61%)0.03 (-7.04%)0.01 (-36.77%)202.70 (+7.59%)169.94 (+50.76%)190.00 (+102.56%)109.70 (+37.64%)37.79 (-13.28%)
1531781 — 2026-01-23 18:49:240.07 (n/a)0.05 (n/a)0.06 (n/a)0.03 (n/a)0.01 (n/a)188.40 (n/a)112.72 (n/a)93.80 (n/a)79.70 (n/a)43.58 (n/a)

dequant_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.04 (-13.65%)0.04 (-13.67%)0.04 (-18.56%)0.03 (-13.52%)0.00 (-23.88%)170.40 (+15.68%)140.12 (+15.33%)135.80 (+22.78%)119.60 (+15.78%)17.94 (+1.87%)
0a6c11c — 2025-12-04 00:39:100.05 (n/a)0.04 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)147.30 (n/a)121.50 (n/a)110.60 (n/a)103.30 (n/a)17.61 (n/a)

dequant_8_cols_1_channels_2048_tile_256_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.04 (-9.01%)0.03 (-16.43%)0.03 (-19.38%)0.03 (-18.29%)0.01 (+13.95%)204.30 (+22.41%)167.36 (+20.70%)168.60 (+24.06%)129.20 (+9.86%)26.68 (+48.68%)
1531781 — 2026-01-23 18:49:240.04 (n/a)0.04 (n/a)0.04 (n/a)0.03 (n/a)0.00 (n/a)166.90 (n/a)138.66 (n/a)135.90 (n/a)117.60 (n/a)17.94 (n/a)

dequant_8_cols_2_channels_2048_tile_128

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.05 (+17.21%)0.03 (-1.70%)0.03 (-3.04%)0.03 (-16.20%)0.01 (+95.59%)198.20 (+19.33%)160.00 (+4.92%)166.20 (+3.17%)106.50 (-14.66%)29.86 (+87.19%)
0a6c11c — 2025-12-04 00:39:100.04 (n/a)0.03 (n/a)0.03 (n/a)0.03 (n/a)0.00 (n/a)166.10 (n/a)152.50 (n/a)161.10 (n/a)124.80 (n/a)15.95 (n/a)

dequant_8_cols_2_channels_2048_tile_128_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.03 (-35.93%)0.03 (-32.34%)0.03 (-36.56%)0.02 (+0.02%)0.00 (-66.99%)229.80 (-0.04%)202.86 (+39.33%)206.50 (+57.63%)165.70 (+56.03%)23.51 (-52.06%)
1531781 — 2026-01-23 18:49:240.05 (n/a)0.04 (n/a)0.04 (n/a)0.02 (n/a)0.01 (n/a)229.90 (n/a)145.60 (n/a)131.00 (n/a)106.20 (n/a)49.03 (n/a)

eltwise_add_1_cols_2_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.12 (-28.68%)0.10 (-15.86%)0.10 (+1.27%)0.08 (+6.27%)0.02 (-59.97%)146.60 (-5.91%)122.52 (+11.61%)119.00 (-1.24%)103.50 (+40.24%)19.09 (-44.48%)
1531781 — 2026-01-23 18:49:240.17 (n/a)0.12 (n/a)0.10 (n/a)0.08 (n/a)0.04 (n/a)155.80 (n/a)109.78 (n/a)120.50 (n/a)73.80 (n/a)34.38 (n/a)

eltwise_add_2_cols_2_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.11 (-25.97%)0.09 (-28.14%)0.10 (-26.34%)0.06 (-29.76%)0.02 (-10.19%)208.90 (+42.40%)143.40 (+41.70%)124.50 (+35.77%)109.90 (+35.01%)42.01 (+61.63%)
1531781 — 2026-01-23 18:49:240.15 (n/a)0.13 (n/a)0.13 (n/a)0.08 (n/a)0.03 (n/a)146.70 (n/a)101.20 (n/a)91.70 (n/a)81.40 (n/a)26.00 (n/a)

eltwise_add_4_cols_2_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.10 (-41.46%)0.08 (-45.57%)0.07 (-51.45%)0.07 (-34.67%)0.01 (-50.12%)177.60 (+53.10%)157.30 (+81.93%)166.70 (+106.06%)124.20 (+70.84%)21.71 (+26.05%)
1531781 — 2026-01-23 18:49:240.17 (n/a)0.15 (n/a)0.15 (n/a)0.11 (n/a)0.02 (n/a)116.00 (n/a)86.46 (n/a)80.90 (n/a)72.70 (n/a)17.22 (n/a)

eltwise_add_8_cols_2_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.09 (-42.05%)0.07 (-49.03%)0.07 (-54.24%)0.06 (-43.66%)0.01 (-32.75%)194.30 (+77.44%)177.62 (+97.36%)187.90 (+118.49%)129.90 (+72.51%)26.98 (+100.80%)
1531781 — 2026-01-23 18:49:240.16 (n/a)0.14 (n/a)0.14 (n/a)0.11 (n/a)0.02 (n/a)109.50 (n/a)90.00 (n/a)86.00 (n/a)75.30 (n/a)13.44 (n/a)

eltwise_mul_1_cols_2_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.12 (-12.21%)0.10 (-20.29%)0.12 (-12.93%)0.06 (-42.27%)0.03 (+92.08%)204.60 (+73.24%)130.12 (+32.94%)106.30 (+14.79%)103.00 (+13.94%)43.27 (+274.66%)
1531781 — 2026-01-23 18:49:240.14 (n/a)0.13 (n/a)0.13 (n/a)0.10 (n/a)0.01 (n/a)118.10 (n/a)97.88 (n/a)92.60 (n/a)90.40 (n/a)11.55 (n/a)

eltwise_mul_2_cols_2_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.11 (-16.58%)0.10 (-21.76%)0.10 (-21.39%)0.07 (-24.97%)0.02 (-1.55%)173.30 (+33.31%)131.84 (+29.13%)118.70 (+27.22%)109.20 (+19.87%)26.24 (+58.72%)
1531781 — 2026-01-23 18:49:240.13 (n/a)0.12 (n/a)0.13 (n/a)0.09 (n/a)0.02 (n/a)130.00 (n/a)102.10 (n/a)93.30 (n/a)91.10 (n/a)16.53 (n/a)

eltwise_mul_4_cols_2_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.11 (-34.49%)0.09 (-40.31%)0.09 (-43.06%)0.06 (-41.10%)0.02 (-13.89%)193.60 (+69.82%)146.96 (+70.53%)143.10 (+75.58%)113.20 (+52.56%)33.54 (+111.57%)
1531781 — 2026-01-23 18:49:240.17 (n/a)0.15 (n/a)0.15 (n/a)0.11 (n/a)0.02 (n/a)114.00 (n/a)86.18 (n/a)81.50 (n/a)74.20 (n/a)15.85 (n/a)

eltwise_mul_8_cols_2_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.07 (-41.77%)0.06 (-45.18%)0.06 (-46.84%)0.04 (-50.92%)0.01 (-26.40%)315.40 (+103.75%)209.88 (+86.93%)196.70 (+88.05%)164.00 (+71.73%)61.40 (+154.40%)
1531781 — 2026-01-23 18:49:240.13 (n/a)0.11 (n/a)0.12 (n/a)0.08 (n/a)0.02 (n/a)154.80 (n/a)112.28 (n/a)104.60 (n/a)95.50 (n/a)24.14 (n/a)

gelu_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.07 (-42.41%)0.06 (-44.49%)0.05 (-52.15%)0.05 (-43.16%)0.01 (-58.87%)178.80 (+75.98%)149.06 (+77.88%)151.40 (+109.12%)125.40 (+73.68%)19.94 (+26.58%)
1531781 — 2026-01-23 18:49:240.11 (n/a)0.10 (n/a)0.11 (n/a)0.08 (n/a)0.02 (n/a)101.60 (n/a)83.80 (n/a)72.40 (n/a)72.20 (n/a)15.75 (n/a)

gelu_1_cols_2_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.07 (-41.84%)0.06 (-48.00%)0.05 (-56.77%)0.04 (-42.71%)0.01 (-43.92%)191.90 (+74.45%)154.00 (+91.88%)159.70 (+131.11%)114.20 (+71.99%)31.16 (+68.12%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.11 (n/a)0.12 (n/a)0.07 (n/a)0.02 (n/a)110.00 (n/a)80.26 (n/a)69.10 (n/a)66.40 (n/a)18.54 (n/a)

gelu_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.05 (-55.10%)0.04 (-57.13%)0.05 (-61.39%)0.03 (-41.16%)0.01 (-72.75%)234.10 (+69.88%)188.08 (+121.22%)176.70 (+159.09%)150.90 (+122.57%)31.58 (+4.43%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.10 (n/a)0.12 (n/a)0.06 (n/a)0.03 (n/a)137.80 (n/a)85.02 (n/a)68.20 (n/a)67.80 (n/a)30.24 (n/a)

gelu_2_cols_2_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.06 (-51.08%)0.05 (-57.83%)0.04 (-64.77%)0.04 (-52.83%)0.01 (-48.68%)228.60 (+112.06%)182.66 (+137.84%)197.40 (+184.03%)137.10 (+104.32%)37.18 (+113.71%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.11 (n/a)0.12 (n/a)0.08 (n/a)0.02 (n/a)107.80 (n/a)76.80 (n/a)69.50 (n/a)67.10 (n/a)17.40 (n/a)

gelu_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.06 (-51.57%)0.05 (-52.17%)0.05 (-57.07%)0.04 (-29.61%)0.01 (-68.40%)217.40 (+42.00%)165.92 (+93.92%)161.40 (+132.90%)130.20 (+106.67%)33.37 (-12.05%)
1531781 — 2026-01-23 18:49:240.13 (n/a)0.11 (n/a)0.12 (n/a)0.05 (n/a)0.03 (n/a)153.10 (n/a)85.56 (n/a)69.30 (n/a)63.00 (n/a)37.94 (n/a)

gelu_4_cols_2_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.06 (-45.90%)0.05 (-51.42%)0.05 (-55.89%)0.04 (-45.69%)0.01 (-55.82%)203.40 (+84.07%)162.56 (+103.71%)159.70 (+126.52%)129.50 (+85.00%)26.38 (+50.91%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.11 (n/a)0.12 (n/a)0.07 (n/a)0.02 (n/a)110.50 (n/a)79.80 (n/a)70.50 (n/a)70.00 (n/a)17.48 (n/a)

gelu_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.05 (-60.93%)0.04 (-60.88%)0.04 (-68.27%)0.03 (-14.20%)0.01 (-84.43%)242.70 (+16.57%)213.04 (+115.76%)226.50 (+215.02%)179.80 (+156.13%)27.64 (-54.84%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.10 (n/a)0.11 (n/a)0.04 (n/a)0.03 (n/a)208.20 (n/a)98.74 (n/a)71.90 (n/a)70.20 (n/a)61.20 (n/a)

gelu_8_cols_2_channels_2048_tile_128

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.05 (-51.69%)0.04 (-56.26%)0.04 (-61.69%)0.04 (-41.15%)0.00 (-70.76%)226.70 (+69.94%)208.28 (+122.90%)214.80 (+161.00%)169.50 (+106.96%)22.37 (-0.42%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.09 (n/a)0.10 (n/a)0.06 (n/a)0.02 (n/a)133.40 (n/a)93.44 (n/a)82.30 (n/a)81.90 (n/a)22.47 (n/a)

gemm_1792x896x1152_64x32x48_8cols_ccolmaj

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
71486ae — 2026-02-09 15:31:514.85 (-5.89%)4.37 (-10.68%)4.18 (-16.27%)4.11 (-3.89%)0.34 (-5.26%)2291.00 (+4.06%)2160.70 (+11.96%)2251.40 (+19.44%)1941.00 (+6.26%)160.60 (+4.25%)1905.95 (-5.89%)1720.01 (-10.68%)1643.15 (-16.27%)1614.77 (-3.89%)132.61 (-5.26%)
1531781 — 2026-01-23 18:49:245.15 (n/a)4.90 (n/a)4.99 (n/a)4.27 (n/a)0.36 (n/a)2201.70 (n/a)1929.94 (n/a)1885.00 (n/a)1826.70 (n/a)154.05 (n/a)2025.19 (n/a)1925.76 (n/a)1962.51 (n/a)1680.20 (n/a)139.97 (n/a)

gemm_192x384x64_48x96x16_4cols

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
71486ae — 2026-02-09 15:31:511.17 (-50.60%)1.02 (-44.70%)1.05 (-40.79%)0.89 (-32.72%)0.11 (-71.06%)249.40 (+48.63%)218.34 (+75.74%)210.00 (+68.81%)189.80 (+102.56%)24.53 (-12.74%)49.73 (-50.60%)43.66 (-44.70%)44.93 (-40.79%)37.84 (-32.72%)4.84 (-71.06%)
1531781 — 2026-01-23 18:49:242.36 (n/a)1.85 (n/a)1.78 (n/a)1.32 (n/a)0.39 (n/a)167.80 (n/a)124.24 (n/a)124.40 (n/a)93.70 (n/a)28.11 (n/a)100.67 (n/a)78.94 (n/a)75.88 (n/a)56.25 (n/a)16.73 (n/a)

gemm_192x384x64_48x96x16_4cols_bcolmaj_ccolmaj

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
71486ae — 2026-02-09 15:31:511.82 (-17.91%)1.38 (-33.12%)1.25 (-43.20%)0.94 (-41.46%)0.38 (+44.05%)236.00 (+70.77%)171.08 (+56.67%)176.90 (+76.02%)121.60 (+21.84%)47.41 (+186.65%)77.61 (-17.91%)58.70 (-33.12%)53.35 (-43.20%)39.98 (-41.46%)16.22 (+44.05%)
1531781 — 2026-01-23 18:49:242.22 (n/a)2.06 (n/a)2.20 (n/a)1.60 (n/a)0.26 (n/a)138.20 (n/a)109.20 (n/a)100.50 (n/a)99.80 (n/a)16.54 (n/a)94.54 (n/a)87.77 (n/a)93.91 (n/a)68.30 (n/a)11.26 (n/a)

gemm_2048x2048x2048_64x64x32_8_cols_0_bcolmaj_0_ccolmaj_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)GFLOP/s (max)GFLOP/s (mean)GFLOP/s (median)GFLOP/s (min)GFLOP/s (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3a053d3 — 2025-11-26 01:12:163.72 (-0.11%)3.55 (-2.75%)3.56 (-2.75%)3.36 (-5.33%)0.12 (+76.10%)2725.23 (-0.11%)2602.52 (-2.75%)2612.11 (-2.75%)2459.89 (-5.33%)86.27 (+76.10%)6984.00 (+5.63%)6608.60 (+2.91%)6577.00 (+2.83%)6304.00 (+0.11%)221.87 (+87.15%)
9e98049 — 2025-11-25 20:50:553.72 (n/a)3.65 (n/a)3.66 (n/a)3.54 (n/a)0.07 (n/a)2728.26 (n/a)2676.06 (n/a)2686.03 (n/a)2598.29 (n/a)48.99 (n/a)6612.00 (n/a)6422.00 (n/a)6396.00 (n/a)6297.00 (n/a)118.55 (n/a)

gemm_2048x2048x2048_64x64x32_8_cols_0_bcolmaj_1_ccolmaj_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)GFLOP/s (max)GFLOP/s (mean)GFLOP/s (median)GFLOP/s (min)GFLOP/s (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3a053d3 — 2025-11-26 01:12:163.40 (-3.14%)3.38 (-0.82%)3.38 (-0.82%)3.35 (+4.93%)0.02 (-84.51%)2493.81 (-3.14%)2475.27 (-0.82%)2479.06 (-0.82%)2453.92 (+4.93%)13.33 (-84.51%)7001.00 (-4.70%)6940.80 (+0.71%)6930.00 (+0.83%)6889.00 (+3.24%)37.43 (-84.76%)
9e98049 — 2025-11-25 20:50:553.51 (n/a)3.40 (n/a)3.41 (n/a)3.19 (n/a)0.12 (n/a)2574.53 (n/a)2495.72 (n/a)2499.62 (n/a)2338.67 (n/a)86.02 (n/a)7346.00 (n/a)6892.20 (n/a)6873.00 (n/a)6673.00 (n/a)245.57 (n/a)

gemm_2048x2048x2048_64x64x32_8_cols_1_bcolmaj_0_ccolmaj_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)GFLOP/s (max)GFLOP/s (mean)GFLOP/s (median)GFLOP/s (min)GFLOP/s (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3a053d3 — 2025-11-26 01:12:163.02 (+0.52%)2.87 (-2.05%)2.86 (-2.24%)2.70 (-6.16%)0.11 (+131.66%)2216.19 (+0.52%)2104.44 (-2.05%)2094.08 (-2.24%)1980.84 (-6.16%)80.25 (+131.67%)8673.00 (+6.56%)8175.60 (+2.21%)8204.00 (+2.29%)7752.00 (-0.51%)314.03 (+145.36%)
9e98049 — 2025-11-25 20:50:553.01 (n/a)2.93 (n/a)2.92 (n/a)2.88 (n/a)0.05 (n/a)2204.81 (n/a)2148.41 (n/a)2142.13 (n/a)2110.81 (n/a)34.64 (n/a)8139.00 (n/a)7998.60 (n/a)8020.00 (n/a)7792.00 (n/a)127.99 (n/a)

gemm_2048x2048x2048_64x64x64_1cols

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
71486ae — 2026-02-09 15:31:510.52 (-0.36%)0.52 (-0.39%)0.52 (-0.43%)0.52 (-0.37%)0.00 (+12.01%)48595.20 (+0.37%)48561.24 (+0.39%)48573.30 (+0.44%)48519.80 (+0.36%)33.18 (+12.89%)354.08 (-0.36%)353.78 (-0.39%)353.69 (-0.43%)353.53 (-0.37%)0.24 (+12.02%)
1531781 — 2026-01-23 18:49:240.52 (n/a)0.52 (n/a)0.52 (n/a)0.52 (n/a)0.00 (n/a)48415.10 (n/a)48370.52 (n/a)48362.20 (n/a)48344.40 (n/a)29.40 (n/a)355.36 (n/a)355.17 (n/a)355.23 (n/a)354.85 (n/a)0.22 (n/a)

gemm_2048x2048x2048_64x64x64_2_cols_0_bcolmaj_0_ccolmaj_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:491.04 (+0.94%)1.03 (+0.23%)1.03 (+0.09%)1.02 (+0.07%)0.00 (+197.51%)24576.00 (-0.07%)24485.82 (-0.23%)24530.40 (-0.09%)24266.40 (-0.94%)111.26 (+194.35%)
0a6c11c — 2025-12-04 00:39:101.03 (n/a)1.03 (n/a)1.03 (n/a)1.02 (n/a)0.00 (n/a)24594.30 (n/a)24542.14 (n/a)24551.90 (n/a)24495.50 (n/a)37.80 (n/a)

gemm_2048x2048x2048_64x64x64_2_cols_0_bcolmaj_0_ccolmaj_0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
a4b6ffe — 2026-01-05 19:24:051.04 (+0.28%)1.03 (-0.15%)1.03 (+0.05%)1.02 (-0.52%)0.01 (+42.72%)24719.80 (+0.52%)24512.42 (+0.16%)24529.20 (-0.05%)24092.60 (-0.28%)252.92 (+43.15%)713.08 (+0.28%)700.92 (-0.15%)700.38 (+0.05%)694.98 (-0.52%)7.30 (+42.71%)
50ad37c — 2025-12-29 20:35:531.04 (n/a)1.03 (n/a)1.03 (n/a)1.02 (n/a)0.01 (n/a)24592.30 (n/a)24473.70 (n/a)24540.90 (n/a)24160.80 (n/a)176.68 (n/a)711.06 (n/a)702.00 (n/a)700.05 (n/a)698.59 (n/a)5.12 (n/a)

gemm_2048x2048x2048_64x64x64_2_cols_0_bcolmaj_1_ccolmaj_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.99 (+1.69%)0.97 (+0.29%)0.97 (-0.04%)0.97 (-0.15%)0.01 (+434.92%)26025.30 (+0.15%)25839.34 (-0.29%)25916.70 (+0.04%)25446.00 (-1.67%)202.05 (+425.44%)
0a6c11c — 2025-12-04 00:39:100.97 (n/a)0.97 (n/a)0.97 (n/a)0.97 (n/a)0.00 (n/a)25986.90 (n/a)25913.52 (n/a)25905.70 (n/a)25877.00 (n/a)38.45 (n/a)

gemm_2048x2048x2048_64x64x64_2_cols_0_bcolmaj_1_ccolmaj_0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
a4b6ffe — 2026-01-05 19:24:050.97 (-0.01%)0.97 (+0.03%)0.97 (+0.04%)0.97 (+0.00%)0.00 (-6.43%)25927.90 (-0.00%)25903.20 (-0.03%)25898.60 (-0.04%)25892.80 (+0.01%)14.05 (-6.56%)663.50 (-0.01%)663.23 (+0.03%)663.35 (+0.04%)662.60 (+0.00%)0.36 (-6.43%)
50ad37c — 2025-12-29 20:35:530.97 (n/a)0.97 (n/a)0.97 (n/a)0.97 (n/a)0.00 (n/a)25928.10 (n/a)25910.14 (n/a)25908.10 (n/a)25891.20 (n/a)15.04 (n/a)663.54 (n/a)663.06 (n/a)663.11 (n/a)662.60 (n/a)0.38 (n/a)

gemm_2048x2048x2048_64x64x64_2_cols_1_bcolmaj_0_ccolmaj_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.92 (-1.02%)0.92 (-0.66%)0.92 (-0.10%)0.91 (-0.78%)0.01 (-4.54%)27751.40 (+0.79%)27480.74 (+0.66%)27354.80 (+0.10%)27320.50 (+1.03%)182.69 (-2.84%)
0a6c11c — 2025-12-04 00:39:100.93 (n/a)0.92 (n/a)0.92 (n/a)0.91 (n/a)0.01 (n/a)27534.80 (n/a)27299.32 (n/a)27326.70 (n/a)27042.30 (n/a)188.04 (n/a)

gemm_2048x2048x2048_64x64x64_2_cols_1_bcolmaj_0_ccolmaj_0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
a4b6ffe — 2026-01-05 19:24:050.92 (+0.11%)0.91 (-0.12%)0.91 (-0.21%)0.91 (-0.24%)0.00 (+30.34%)27723.50 (+0.24%)27548.94 (+0.12%)27522.90 (+0.21%)27403.60 (-0.11%)124.26 (+30.58%)626.92 (+0.11%)623.62 (-0.12%)624.20 (-0.21%)619.69 (-0.24%)2.81 (+30.34%)
50ad37c — 2025-12-29 20:35:530.92 (n/a)0.91 (n/a)0.92 (n/a)0.91 (n/a)0.00 (n/a)27657.10 (n/a)27515.58 (n/a)27464.70 (n/a)27434.60 (n/a)95.16 (n/a)626.21 (n/a)624.37 (n/a)625.52 (n/a)621.17 (n/a)2.16 (n/a)

gemm_2048x2048x2048_64x64x64_2cols_bcolmaj

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
71486ae — 2026-02-09 15:31:510.91 (-1.68%)0.91 (-0.57%)0.91 (-0.65%)0.90 (+0.04%)0.00 (-60.67%)27816.10 (-0.04%)27688.98 (+0.56%)27701.10 (+0.66%)27568.60 (+1.71%)108.99 (-59.98%)623.17 (-1.68%)620.47 (-0.57%)620.19 (-0.65%)617.62 (+0.04%)2.44 (-60.67%)
1531781 — 2026-01-23 18:49:240.93 (n/a)0.91 (n/a)0.91 (n/a)0.90 (n/a)0.01 (n/a)27826.70 (n/a)27533.84 (n/a)27519.90 (n/a)27105.20 (n/a)272.35 (n/a)633.82 (n/a)624.00 (n/a)624.27 (n/a)617.39 (n/a)6.21 (n/a)

gemm_2048x2048x2048_64x64x64_8_cols_0_bcolmaj_0_ccolmaj_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)GFLOP/s (max)GFLOP/s (mean)GFLOP/s (median)GFLOP/s (min)GFLOP/s (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
31ab24e — 2025-11-22 02:01:563.65 (-1.06%)3.59 (+0.83%)3.63 (-0.26%)3.45 (+2.66%)0.07 (-44.71%)2678.08 (-1.06%)2632.13 (+0.83%)2658.60 (-0.26%)2528.31 (+2.66%)53.86 (-44.71%)6795.00 (-2.59%)6529.80 (-0.92%)6462.00 (+0.26%)6415.00 (+1.07%)137.14 (-45.27%)
d9a64c5 — 2025-11-21 20:49:353.69 (n/a)3.56 (n/a)3.64 (n/a)3.36 (n/a)0.13 (n/a)2706.77 (n/a)2610.50 (n/a)2665.61 (n/a)2462.71 (n/a)97.41 (n/a)6976.00 (n/a)6590.40 (n/a)6445.00 (n/a)6347.00 (n/a)250.58 (n/a)

gemm_2048x2048x2048_64x64x64_8_cols_0_bcolmaj_0_ccolmaj_0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
d48746f — 2026-01-12 22:17:103.98 (+0.20%)3.84 (-2.95%)3.76 (-5.27%)3.72 (-4.97%)0.13 (+473.97%)6760.00 (+5.23%)6559.50 (+3.13%)6697.10 (+5.56%)6321.00 (-0.20%)218.68 (+499.40%)2717.91 (+0.20%)2621.44 (-2.95%)2565.27 (-5.27%)2541.40 (-4.97%)88.37 (+473.97%)
331dcca — 2026-01-08 18:34:103.97 (n/a)3.96 (n/a)3.97 (n/a)3.92 (n/a)0.02 (n/a)6423.80 (n/a)6360.32 (n/a)6344.30 (n/a)6333.90 (n/a)36.48 (n/a)2712.39 (n/a)2701.17 (n/a)2707.91 (n/a)2674.40 (n/a)15.40 (n/a)

gemm_2048x2048x2048_64x64x64_8_cols_0_bcolmaj_1_ccolmaj_0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
d48746f — 2026-01-12 22:17:103.75 (+0.28%)3.70 (+1.25%)3.71 (-0.21%)3.65 (+4.77%)0.04 (-66.90%)6901.80 (-4.56%)6795.46 (-1.30%)6790.20 (+0.21%)6715.90 (-0.28%)67.05 (-68.40%)2558.08 (+0.28%)2528.33 (+1.25%)2530.10 (-0.21%)2489.18 (+4.77%)24.81 (-66.90%)
331dcca — 2026-01-08 18:34:103.74 (n/a)3.66 (n/a)3.71 (n/a)3.48 (n/a)0.11 (n/a)7231.20 (n/a)6885.18 (n/a)6775.70 (n/a)6734.90 (n/a)212.21 (n/a)2550.87 (n/a)2497.04 (n/a)2535.50 (n/a)2375.78 (n/a)74.96 (n/a)

gemm_2048x2048x2048_64x64x64_8_cols_1_bcolmaj_0_ccolmaj_0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
d48746f — 2026-01-12 22:17:103.62 (-0.77%)3.54 (+0.14%)3.60 (-0.04%)3.44 (+2.17%)0.09 (-25.77%)7307.40 (-2.13%)7103.46 (-0.18%)6982.50 (+0.04%)6948.20 (+0.77%)182.95 (-26.55%)2472.55 (-0.77%)2419.80 (+0.14%)2460.41 (-0.04%)2351.04 (+2.17%)61.76 (-25.77%)
331dcca — 2026-01-08 18:34:103.65 (n/a)3.54 (n/a)3.61 (n/a)3.37 (n/a)0.12 (n/a)7466.20 (n/a)7116.40 (n/a)6979.60 (n/a)6895.10 (n/a)249.09 (n/a)2491.61 (n/a)2416.45 (n/a)2461.45 (n/a)2301.00 (n/a)83.20 (n/a)

gemm_2048x2048x2048_64x64x64_8cols_bcolmaj_ccolmaj

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
71486ae — 2026-02-09 15:31:513.36 (-2.87%)3.27 (-4.28%)3.29 (-4.42%)3.16 (-3.92%)0.08 (+10.34%)7963.20 (+4.08%)7698.64 (+4.48%)7638.50 (+4.63%)7478.70 (+2.96%)192.12 (+18.21%)2297.17 (-2.87%)2232.65 (-4.28%)2249.13 (-4.42%)2157.41 (-3.92%)55.34 (+10.34%)
1531781 — 2026-01-23 18:49:243.46 (n/a)3.42 (n/a)3.45 (n/a)3.29 (n/a)0.07 (n/a)7651.20 (n/a)7368.34 (n/a)7300.80 (n/a)7263.70 (n/a)162.53 (n/a)2365.17 (n/a)2332.46 (n/a)2353.14 (n/a)2245.39 (n/a)50.15 (n/a)

gemm_384x1536x1792_32x48x64_4cols_bcolmaj

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
71486ae — 2026-02-09 15:31:514.19 (-3.78%)3.85 (-9.17%)3.88 (-10.47%)3.51 (-8.76%)0.27 (+22.49%)2296.00 (+9.60%)2101.42 (+10.27%)2079.70 (+11.69%)1924.80 (+3.93%)147.56 (+39.36%)1098.29 (-3.78%)1009.90 (-9.17%)1016.45 (-10.47%)920.71 (-8.76%)70.45 (+22.49%)
1531781 — 2026-01-23 18:49:244.35 (n/a)4.24 (n/a)4.33 (n/a)3.85 (n/a)0.22 (n/a)2094.90 (n/a)1905.64 (n/a)1862.00 (n/a)1852.00 (n/a)105.88 (n/a)1141.45 (n/a)1111.86 (n/a)1135.29 (n/a)1009.08 (n/a)57.51 (n/a)

gemm_896x1792x640_32x64x80_8cols_ccolmaj

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
71486ae — 2026-02-09 15:31:514.75 (-30.09%)4.65 (-27.76%)4.68 (-30.16%)4.47 (-16.16%)0.12 (-80.81%)1487.50 (+19.28%)1432.10 (+37.35%)1421.70 (+43.19%)1399.50 (+43.05%)37.04 (-67.67%)1468.51 (-30.09%)1435.87 (-27.76%)1445.61 (-30.16%)1381.66 (-16.16%)36.63 (-80.81%)
1531781 — 2026-01-23 18:49:246.80 (n/a)6.43 (n/a)6.70 (n/a)5.33 (n/a)0.62 (n/a)1247.10 (n/a)1042.70 (n/a)992.90 (n/a)978.30 (n/a)114.59 (n/a)2100.72 (n/a)1987.76 (n/a)2069.85 (n/a)1647.93 (n/a)190.83 (n/a)

layer_norm_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.07 (-39.66%)0.06 (-33.25%)0.06 (-31.32%)0.05 (-25.04%)0.01 (-64.06%)159.70 (+33.42%)141.28 (+44.99%)146.40 (+45.67%)118.50 (+65.73%)17.79 (-21.54%)
1531781 — 2026-01-23 18:49:240.11 (n/a)0.09 (n/a)0.08 (n/a)0.07 (n/a)0.02 (n/a)119.70 (n/a)97.44 (n/a)100.50 (n/a)71.50 (n/a)22.67 (n/a)

layer_norm_1_cols_2_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.07 (-33.64%)0.05 (-40.59%)0.05 (-46.34%)0.03 (-42.63%)0.02 (-18.65%)240.80 (+74.37%)171.50 (+73.62%)181.00 (+86.41%)115.20 (+50.79%)51.90 (+108.58%)
1531781 — 2026-01-23 18:49:240.11 (n/a)0.09 (n/a)0.08 (n/a)0.06 (n/a)0.02 (n/a)138.10 (n/a)98.78 (n/a)97.10 (n/a)76.40 (n/a)24.88 (n/a)

layer_norm_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.06 (-52.49%)0.05 (-43.41%)0.05 (-37.46%)0.04 (-33.73%)0.01 (-60.61%)211.50 (+50.96%)169.32 (+72.11%)150.90 (+59.85%)143.30 (+110.43%)31.83 (+21.93%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.09 (n/a)0.09 (n/a)0.06 (n/a)0.02 (n/a)140.10 (n/a)98.38 (n/a)94.40 (n/a)68.10 (n/a)26.11 (n/a)

layer_norm_2_cols_2_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.06 (-51.20%)0.06 (-40.01%)0.06 (-31.55%)0.04 (-29.24%)0.01 (-70.37%)184.30 (+41.33%)146.58 (+59.50%)136.60 (+46.10%)127.10 (+105.00%)22.48 (-12.14%)
1531781 — 2026-01-23 18:49:240.13 (n/a)0.09 (n/a)0.09 (n/a)0.06 (n/a)0.03 (n/a)130.40 (n/a)91.90 (n/a)93.50 (n/a)62.00 (n/a)25.59 (n/a)

layer_norm_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.07 (-39.21%)0.05 (-45.74%)0.05 (-49.11%)0.04 (-36.01%)0.01 (-45.93%)220.30 (+56.24%)166.18 (+81.54%)153.10 (+96.53%)120.20 (+64.66%)39.04 (+37.34%)
1531781 — 2026-01-23 18:49:240.11 (n/a)0.09 (n/a)0.11 (n/a)0.06 (n/a)0.02 (n/a)141.00 (n/a)91.54 (n/a)77.90 (n/a)73.00 (n/a)28.43 (n/a)

layer_norm_4_cols_2_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.07 (-38.63%)0.05 (-42.03%)0.04 (-52.24%)0.04 (-27.52%)0.01 (-52.70%)204.80 (+38.01%)171.66 (+66.43%)183.80 (+109.34%)120.20 (+63.09%)33.67 (+4.27%)
1531781 — 2026-01-23 18:49:240.11 (n/a)0.09 (n/a)0.09 (n/a)0.06 (n/a)0.02 (n/a)148.40 (n/a)103.14 (n/a)87.80 (n/a)73.70 (n/a)32.30 (n/a)

layer_norm_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.07 (-45.61%)0.05 (-49.05%)0.05 (-50.36%)0.03 (-37.56%)0.01 (-55.46%)246.50 (+60.17%)170.74 (+88.96%)155.90 (+101.42%)123.70 (+83.80%)45.94 (+28.10%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.10 (n/a)0.11 (n/a)0.05 (n/a)0.03 (n/a)153.90 (n/a)90.36 (n/a)77.40 (n/a)67.30 (n/a)35.86 (n/a)

layer_norm_8_cols_2_channels_2048_tile_128

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.06 (-30.93%)0.05 (-32.42%)0.06 (-32.86%)0.04 (-25.78%)0.01 (-33.81%)189.10 (+34.78%)157.08 (+47.49%)145.40 (+48.98%)133.80 (+44.81%)25.43 (+28.17%)
1531781 — 2026-01-23 18:49:240.09 (n/a)0.08 (n/a)0.08 (n/a)0.06 (n/a)0.01 (n/a)140.30 (n/a)106.50 (n/a)97.60 (n/a)92.40 (n/a)19.84 (n/a)

matrix_vector_mul_128x128_32_1col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:490.37 (+17.13%)0.30 (+24.87%)0.30 (+38.03%)0.23 (+12.47%)0.06 (+35.23%)
0a6c11c — 2025-12-04 00:39:100.32 (n/a)0.24 (n/a)0.22 (n/a)0.20 (n/a)0.04 (n/a)

matrix_vector_mul_128x128_32_1col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:100.26 (-1.81%)0.24 (+4.06%)0.24 (+0.52%)0.21 (+16.97%)0.02 (-48.16%)0.26 (-1.81%)0.24 (+4.06%)0.24 (+0.52%)0.20 (+16.97%)0.02 (-48.16%)
a4b6ffe — 2026-01-05 19:24:050.26 (n/a)0.23 (n/a)0.24 (n/a)0.18 (n/a)0.04 (n/a)0.26 (n/a)0.23 (n/a)0.24 (n/a)0.17 (n/a)0.04 (n/a)

matrix_vector_mul_128x128_32tsi_128tso_1col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
71486ae — 2026-02-09 15:31:510.28 (-36.51%)0.23 (-32.83%)0.21 (-39.93%)0.19 (-12.40%)0.05 (-44.42%)0.28 (-36.51%)0.23 (-32.83%)0.21 (-39.93%)0.19 (-12.40%)0.04 (-44.42%)
1531781 — 2026-01-23 18:49:240.45 (n/a)0.35 (n/a)0.35 (n/a)0.22 (n/a)0.08 (n/a)0.44 (n/a)0.34 (n/a)0.34 (n/a)0.22 (n/a)0.08 (n/a)

matrix_vector_mul_2048x8192_1_1col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:4913.57 (+3.47%)13.15 (+1.09%)13.10 (+0.16%)12.52 (-1.58%)0.37 (+153.19%)
0a6c11c — 2025-12-04 00:39:1013.11 (n/a)13.01 (n/a)13.07 (n/a)12.72 (n/a)0.15 (n/a)

matrix_vector_mul_2048x8192_1_1col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:1013.43 (-0.07%)13.30 (+4.62%)13.27 (+1.29%)13.16 (+16.70%)0.12 (-86.69%)13.42 (-0.07%)13.29 (+4.62%)13.26 (+1.29%)13.16 (+16.70%)0.12 (-86.69%)
a4b6ffe — 2026-01-05 19:24:0513.44 (n/a)12.71 (n/a)13.10 (n/a)11.28 (n/a)0.87 (n/a)13.43 (n/a)12.70 (n/a)13.09 (n/a)11.27 (n/a)0.87 (n/a)

matrix_vector_mul_2048x8192_1_2col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:4925.06 (+2.47%)23.71 (+1.81%)24.13 (+0.50%)20.90 (+5.41%)1.48 (-15.60%)
0a6c11c — 2025-12-04 00:39:1024.46 (n/a)23.29 (n/a)24.01 (n/a)19.83 (n/a)1.75 (n/a)

matrix_vector_mul_2048x8192_1_2col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:1024.46 (-0.60%)21.07 (-8.03%)19.75 (-17.83%)18.08 (+1.18%)3.03 (+7.07%)24.44 (-0.60%)21.06 (-8.03%)19.74 (-17.83%)18.07 (+1.18%)3.03 (+7.07%)
a4b6ffe — 2026-01-05 19:24:0524.60 (n/a)22.91 (n/a)24.04 (n/a)17.87 (n/a)2.83 (n/a)24.59 (n/a)22.90 (n/a)24.02 (n/a)17.86 (n/a)2.83 (n/a)

matrix_vector_mul_2048x8192_1_4col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:4942.05 (+1.54%)40.50 (+12.60%)40.49 (+2.36%)38.98 (+97.53%)0.97 (-88.09%)
0a6c11c — 2025-12-04 00:39:1041.41 (n/a)35.97 (n/a)39.55 (n/a)19.73 (n/a)8.15 (n/a)

matrix_vector_mul_2048x8192_1_4col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:1042.58 (+1.51%)41.37 (+14.29%)41.25 (+4.89%)40.33 (+91.42%)0.92 (-89.18%)42.56 (+1.51%)41.34 (+14.29%)41.23 (+4.89%)40.30 (+91.42%)0.92 (-89.18%)
a4b6ffe — 2026-01-05 19:24:0541.95 (n/a)36.20 (n/a)39.33 (n/a)21.07 (n/a)8.54 (n/a)41.92 (n/a)36.17 (n/a)39.31 (n/a)21.05 (n/a)8.53 (n/a)

matrix_vector_mul_2048x8192_1_8col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:4946.62 (+3.11%)44.82 (+0.17%)44.78 (+0.17%)42.60 (-3.80%)1.60 (+367.72%)
0a6c11c — 2025-12-04 00:39:1045.21 (n/a)44.75 (n/a)44.70 (n/a)44.29 (n/a)0.34 (n/a)

matrix_vector_mul_2048x8192_1_8col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:1046.04 (+4.14%)45.06 (+3.47%)44.55 (+2.76%)44.30 (+2.84%)0.87 (+66.58%)46.01 (+4.14%)45.03 (+3.47%)44.52 (+2.76%)44.27 (+2.84%)0.87 (+66.58%)
a4b6ffe — 2026-01-05 19:24:0544.21 (n/a)43.55 (n/a)43.35 (n/a)43.07 (n/a)0.52 (n/a)44.18 (n/a)43.52 (n/a)43.32 (n/a)43.05 (n/a)0.52 (n/a)

matrix_vector_mul_2048x8192_1tsi_1024tso_2col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
71486ae — 2026-02-09 15:31:5125.23 (+1.71%)24.75 (+0.63%)24.81 (+1.06%)23.93 (-1.94%)0.50 (+183.51%)25.21 (+1.71%)24.73 (+0.63%)24.79 (+1.06%)23.91 (-1.94%)0.50 (+183.51%)
1531781 — 2026-01-23 18:49:2424.80 (n/a)24.59 (n/a)24.55 (n/a)24.40 (n/a)0.18 (n/a)24.79 (n/a)24.58 (n/a)24.53 (n/a)24.38 (n/a)0.18 (n/a)

matrix_vector_mul_2048x8192_1tsi_2048tso_1col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
71486ae — 2026-02-09 15:31:5113.43 (-0.81%)12.93 (-3.10%)13.28 (-1.20%)11.22 (-14.33%)0.96 (+363.08%)13.42 (-0.81%)12.92 (-3.10%)13.28 (-1.20%)11.22 (-14.33%)0.95 (+363.09%)
1531781 — 2026-01-23 18:49:2413.54 (n/a)13.34 (n/a)13.45 (n/a)13.10 (n/a)0.21 (n/a)13.53 (n/a)13.33 (n/a)13.44 (n/a)13.09 (n/a)0.21 (n/a)

matrix_vector_mul_2048x8192_1tsi_256tso_8col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
71486ae — 2026-02-09 15:31:5146.35 (-0.98%)44.57 (-2.91%)44.09 (-4.93%)42.67 (-4.78%)1.50 (+67.94%)46.33 (-0.98%)44.54 (-2.91%)44.06 (-4.93%)42.65 (-4.78%)1.50 (+67.94%)
1531781 — 2026-01-23 18:49:2446.81 (n/a)45.91 (n/a)46.38 (n/a)44.81 (n/a)0.89 (n/a)46.79 (n/a)45.88 (n/a)46.35 (n/a)44.79 (n/a)0.89 (n/a)

matrix_vector_mul_2048x8192_1tsi_512tso_4col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
71486ae — 2026-02-09 15:31:5142.66 (+2.82%)40.93 (+0.28%)40.45 (-1.63%)39.09 (-2.17%)1.52 (+133.79%)42.63 (+2.82%)40.90 (+0.28%)40.43 (-1.63%)39.06 (-2.17%)1.52 (+133.79%)
1531781 — 2026-01-23 18:49:2441.49 (n/a)40.81 (n/a)41.13 (n/a)39.95 (n/a)0.65 (n/a)41.46 (n/a)40.79 (n/a)41.10 (n/a)39.93 (n/a)0.65 (n/a)

matrix_vector_mul_8192x2048_4_1col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:4913.29 (+1.63%)13.19 (+6.08%)13.19 (+1.17%)13.09 (+26.86%)0.08 (-92.94%)
0a6c11c — 2025-12-04 00:39:1013.08 (n/a)12.44 (n/a)13.03 (n/a)10.32 (n/a)1.07 (n/a)

matrix_vector_mul_8192x2048_4_1col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:1013.68 (+3.42%)13.30 (+8.54%)13.38 (+7.25%)12.72 (+14.96%)0.35 (-66.09%)13.67 (+3.42%)13.29 (+8.54%)13.38 (+7.25%)12.71 (+14.96%)0.35 (-66.09%)
a4b6ffe — 2026-01-05 19:24:0513.23 (n/a)12.25 (n/a)12.48 (n/a)11.07 (n/a)1.04 (n/a)13.22 (n/a)12.24 (n/a)12.47 (n/a)11.06 (n/a)1.04 (n/a)

matrix_vector_mul_8192x2048_4_2col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:4925.58 (+4.24%)24.23 (+6.59%)24.03 (+0.29%)23.58 (+37.29%)0.70 (-74.97%)
0a6c11c — 2025-12-04 00:39:1024.54 (n/a)22.73 (n/a)23.96 (n/a)17.18 (n/a)2.79 (n/a)

matrix_vector_mul_8192x2048_4_2col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:1025.08 (+1.32%)24.68 (+13.42%)24.61 (+3.26%)24.47 (+48.32%)0.24 (-93.56%)25.07 (+1.32%)24.67 (+13.42%)24.59 (+3.26%)24.46 (+48.32%)0.24 (-93.56%)
a4b6ffe — 2026-01-05 19:24:0524.76 (n/a)21.76 (n/a)23.83 (n/a)16.50 (n/a)3.74 (n/a)24.74 (n/a)21.75 (n/a)23.82 (n/a)16.49 (n/a)3.73 (n/a)

matrix_vector_mul_8192x2048_4_4col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:4940.94 (-2.19%)39.74 (-1.10%)39.76 (-0.60%)38.17 (-2.14%)1.04 (+4.39%)
0a6c11c — 2025-12-04 00:39:1041.86 (n/a)40.18 (n/a)40.00 (n/a)39.00 (n/a)1.00 (n/a)

matrix_vector_mul_8192x2048_4_4col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:1040.74 (-1.09%)37.16 (-7.15%)40.25 (+1.47%)24.51 (-37.44%)7.07 (+736.13%)40.72 (-1.09%)37.13 (-7.15%)40.23 (+1.47%)24.50 (-37.44%)7.07 (+736.13%)
a4b6ffe — 2026-01-05 19:24:0541.19 (n/a)40.02 (n/a)39.67 (n/a)39.18 (n/a)0.85 (n/a)41.17 (n/a)39.99 (n/a)39.64 (n/a)39.16 (n/a)0.85 (n/a)

matrix_vector_mul_8192x2048_4_8col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:4946.71 (-1.04%)44.91 (-2.34%)44.71 (-2.98%)43.80 (-1.44%)0.96 (+6.93%)
0a6c11c — 2025-12-04 00:39:1047.20 (n/a)45.98 (n/a)46.08 (n/a)44.44 (n/a)0.90 (n/a)

matrix_vector_mul_8192x2048_4_8col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:1046.28 (+5.68%)44.89 (+14.59%)44.74 (+4.26%)44.20 (+62.80%)0.84 (-87.96%)46.25 (+5.68%)44.86 (+14.59%)44.71 (+4.26%)44.17 (+62.80%)0.84 (-87.96%)
a4b6ffe — 2026-01-05 19:24:0543.79 (n/a)39.17 (n/a)42.91 (n/a)27.15 (n/a)6.98 (n/a)43.77 (n/a)39.15 (n/a)42.88 (n/a)27.13 (n/a)6.98 (n/a)

matrix_vector_mul_8192x2048_4tsi_1024tso_1col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
71486ae — 2026-02-09 15:31:5113.67 (+1.53%)13.52 (+1.47%)13.47 (+1.13%)13.36 (+1.19%)0.14 (+20.49%)13.66 (+1.53%)13.51 (+1.47%)13.46 (+1.13%)13.35 (+1.19%)0.14 (+20.49%)
1531781 — 2026-01-23 18:49:2413.46 (n/a)13.33 (n/a)13.32 (n/a)13.20 (n/a)0.11 (n/a)13.45 (n/a)13.32 (n/a)13.31 (n/a)13.19 (n/a)0.11 (n/a)

matrix_vector_mul_8192x2048_4tsi_1024tso_2col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
71486ae — 2026-02-09 15:31:5125.75 (+1.91%)24.48 (-1.09%)24.84 (+0.92%)22.61 (-7.90%)1.17 (+292.94%)25.73 (+1.91%)24.47 (-1.09%)24.82 (+0.92%)22.59 (-7.90%)1.17 (+292.94%)
1531781 — 2026-01-23 18:49:2425.27 (n/a)24.75 (n/a)24.61 (n/a)24.55 (n/a)0.30 (n/a)25.25 (n/a)24.74 (n/a)24.59 (n/a)24.53 (n/a)0.30 (n/a)

matrix_vector_mul_8192x2048_4tsi_1024tso_4col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
71486ae — 2026-02-09 15:31:5141.94 (+2.63%)40.94 (+2.33%)41.41 (+2.41%)39.65 (+2.01%)1.02 (+15.37%)41.91 (+2.63%)40.91 (+2.33%)41.38 (+2.41%)39.62 (+2.01%)1.02 (+15.37%)
1531781 — 2026-01-23 18:49:2440.86 (n/a)40.01 (n/a)40.43 (n/a)38.86 (n/a)0.89 (n/a)40.84 (n/a)39.98 (n/a)40.41 (n/a)38.84 (n/a)0.89 (n/a)

matrix_vector_mul_8192x2048_4tsi_1024tso_8col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
71486ae — 2026-02-09 15:31:5145.89 (-0.42%)44.18 (-2.05%)44.61 (-1.94%)41.87 (-4.60%)1.50 (+54.95%)45.86 (-0.42%)44.16 (-2.05%)44.58 (-1.94%)41.85 (-4.60%)1.50 (+54.95%)
1531781 — 2026-01-23 18:49:2446.08 (n/a)45.11 (n/a)45.49 (n/a)43.89 (n/a)0.97 (n/a)46.05 (n/a)45.08 (n/a)45.46 (n/a)43.87 (n/a)0.97 (n/a)

mem_copy_16_cores_2_chans_2048_tile_128_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.07 (+22.03%)0.06 (+27.86%)0.06 (+35.49%)0.05 (+15.00%)0.01 (+49.69%)173.30 (-13.00%)137.94 (-21.32%)133.60 (-26.19%)123.40 (-18.06%)18.40 (+8.53%)
0a6c11c — 2025-12-04 00:39:100.05 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.00 (n/a)199.20 (n/a)175.32 (n/a)181.00 (n/a)150.60 (n/a)16.96 (n/a)

mem_copy_16_cores_2_chans_2048_tile_128_False0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.04 (-54.61%)0.04 (-50.48%)0.04 (-55.59%)0.03 (-18.81%)0.00 (-82.31%)240.60 (+23.20%)205.32 (+83.95%)195.00 (+125.17%)186.40 (+120.33%)22.72 (-52.32%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.08 (n/a)0.09 (n/a)0.04 (n/a)0.02 (n/a)195.30 (n/a)111.62 (n/a)86.60 (n/a)84.60 (n/a)47.65 (n/a)

mem_copy_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
003b49d — 2025-11-14 19:30:540.05 (-16.99%)0.05 (-12.47%)0.05 (-7.32%)0.04 (-11.17%)0.00 (-29.61%)197.00 (+12.57%)171.00 (+13.70%)164.00 (+7.89%)153.00 (+20.47%)17.49 (-4.32%)
0999a9a — 2025-11-13 19:24:080.06 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)175.00 (n/a)150.40 (n/a)152.00 (n/a)127.00 (n/a)18.28 (n/a)

mem_copy_1_cols_2_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
003b49d — 2025-11-14 19:30:540.06 (+5.56%)0.05 (-1.59%)0.05 (-1.39%)0.03 (-14.07%)0.01 (+25.03%)263.00 (+16.37%)166.00 (+4.53%)144.00 (+1.41%)126.00 (-5.26%)49.38 (+43.17%)
0999a9a — 2025-11-13 19:24:080.06 (n/a)0.05 (n/a)0.05 (n/a)0.03 (n/a)0.01 (n/a)226.00 (n/a)158.80 (n/a)142.00 (n/a)133.00 (n/a)34.49 (n/a)

mem_copy_1_cores_1_chans_2048_tile_2048_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.08 (+10.93%)0.07 (+7.79%)0.07 (+3.55%)0.05 (-1.73%)0.01 (+36.73%)150.20 (+1.76%)121.68 (-6.36%)125.60 (-3.38%)97.90 (-9.85%)18.56 (+21.98%)
0a6c11c — 2025-12-04 00:39:100.08 (n/a)0.06 (n/a)0.06 (n/a)0.06 (n/a)0.01 (n/a)147.60 (n/a)129.94 (n/a)130.00 (n/a)108.60 (n/a)15.21 (n/a)

mem_copy_1_cores_1_chans_2048_tile_2048_False0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.06 (-44.36%)0.05 (-44.75%)0.05 (-45.76%)0.03 (-51.43%)0.01 (-40.91%)243.20 (+105.93%)176.06 (+82.41%)155.70 (+84.48%)148.10 (+79.73%)40.07 (+119.09%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.09 (n/a)0.10 (n/a)0.07 (n/a)0.02 (n/a)118.10 (n/a)96.52 (n/a)84.40 (n/a)82.40 (n/a)18.29 (n/a)

mem_copy_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
003b49d — 2025-11-14 19:30:540.06 (-15.32%)0.05 (-17.09%)0.05 (-23.81%)0.04 (-15.54%)0.01 (-18.18%)193.00 (+18.40%)160.80 (+20.36%)168.00 (+31.25%)124.00 (+18.10%)22.92 (+11.39%)
0999a9a — 2025-11-13 19:24:080.07 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)163.00 (n/a)133.60 (n/a)128.00 (n/a)105.00 (n/a)20.58 (n/a)

mem_copy_2_cols_2_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
003b49d — 2025-11-14 19:30:540.06 (-23.02%)0.05 (-6.98%)0.05 (-2.13%)0.04 (-15.31%)0.01 (-33.29%)196.00 (+18.07%)147.00 (+6.52%)141.00 (+2.17%)126.00 (+29.90%)25.27 (+8.13%)
0999a9a — 2025-11-13 19:24:080.08 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)166.00 (n/a)138.00 (n/a)138.00 (n/a)97.00 (n/a)23.37 (n/a)

mem_copy_2_cores_1_chans_2048_tile_1024_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.10 (+31.37%)0.08 (+29.12%)0.08 (+29.06%)0.07 (+41.80%)0.01 (+0.23%)109.90 (-29.46%)98.30 (-23.38%)100.80 (-22.52%)79.40 (-23.87%)10.54 (-46.17%)
0a6c11c — 2025-12-04 00:39:100.08 (n/a)0.07 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)155.80 (n/a)128.30 (n/a)130.10 (n/a)104.30 (n/a)19.58 (n/a)

mem_copy_2_cores_1_chans_2048_tile_1024_False0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.05 (-55.22%)0.05 (-51.21%)0.04 (-53.58%)0.04 (-41.99%)0.00 (-79.29%)201.90 (+72.42%)181.16 (+96.49%)187.20 (+115.42%)157.10 (+123.15%)18.42 (-21.01%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.09 (n/a)0.09 (n/a)0.07 (n/a)0.02 (n/a)117.10 (n/a)92.20 (n/a)86.90 (n/a)70.40 (n/a)23.32 (n/a)

mem_copy_2_cores_2_chans_2048_tile_1024_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.10 (+25.02%)0.07 (+10.01%)0.07 (+7.53%)0.05 (-1.99%)0.01 (+77.08%)150.10 (+2.04%)114.60 (-7.22%)111.70 (-6.99%)83.90 (-20.02%)21.23 (+43.54%)
0a6c11c — 2025-12-04 00:39:100.08 (n/a)0.07 (n/a)0.07 (n/a)0.06 (n/a)0.01 (n/a)147.10 (n/a)123.52 (n/a)120.10 (n/a)104.90 (n/a)14.79 (n/a)

mem_copy_2_cores_2_chans_2048_tile_1024_False0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.06 (-35.93%)0.05 (-37.43%)0.05 (-41.85%)0.04 (-24.14%)0.01 (-60.56%)184.80 (+31.81%)158.18 (+55.60%)158.30 (+71.88%)129.50 (+56.21%)19.69 (-18.74%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.08 (n/a)0.09 (n/a)0.06 (n/a)0.02 (n/a)140.20 (n/a)101.66 (n/a)92.10 (n/a)82.90 (n/a)24.23 (n/a)

mem_copy_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
003b49d — 2025-11-14 19:30:540.07 (+11.97%)0.05 (+3.09%)0.05 (+2.10%)0.03 (-11.35%)0.01 (+41.65%)229.00 (+12.81%)156.00 (-0.51%)143.00 (-2.05%)117.00 (-10.69%)38.21 (+48.71%)
0999a9a — 2025-11-13 19:24:080.06 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)203.00 (n/a)156.80 (n/a)146.00 (n/a)131.00 (n/a)25.69 (n/a)

mem_copy_4_cols_2_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
003b49d — 2025-11-14 19:30:540.06 (+9.02%)0.05 (+5.59%)0.05 (+0.00%)0.04 (+3.63%)0.01 (+25.48%)193.00 (-3.50%)147.40 (-4.66%)147.00 (+0.00%)122.00 (-8.27%)25.24 (+7.48%)
0999a9a — 2025-11-13 19:24:080.06 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)200.00 (n/a)154.60 (n/a)147.00 (n/a)133.00 (n/a)23.48 (n/a)

mem_copy_4_cores_1_chans_2048_tile_512_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.09 (-3.98%)0.07 (+4.00%)0.07 (+7.04%)0.06 (+3.67%)0.01 (-15.76%)129.70 (-3.57%)113.30 (-4.50%)116.90 (-6.55%)90.40 (+4.15%)15.04 (-13.80%)
0a6c11c — 2025-12-04 00:39:100.09 (n/a)0.07 (n/a)0.07 (n/a)0.06 (n/a)0.01 (n/a)134.50 (n/a)118.64 (n/a)125.10 (n/a)86.80 (n/a)17.45 (n/a)

mem_copy_4_cores_1_chans_2048_tile_512_False0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.09 (-30.27%)0.06 (-44.91%)0.05 (-53.75%)0.04 (-33.87%)0.02 (-32.03%)194.10 (+51.29%)153.12 (+81.08%)154.70 (+116.06%)94.20 (+43.38%)37.72 (+42.79%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.10 (n/a)0.11 (n/a)0.06 (n/a)0.03 (n/a)128.30 (n/a)84.56 (n/a)71.60 (n/a)65.70 (n/a)26.42 (n/a)

mem_copy_4_cores_2_chans_2048_tile_512_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.08 (+9.57%)0.07 (+16.85%)0.07 (+19.58%)0.06 (+24.52%)0.01 (+1.05%)134.50 (-19.70%)116.64 (-14.92%)116.80 (-16.39%)98.00 (-8.75%)15.71 (-25.13%)
0a6c11c — 2025-12-04 00:39:100.08 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)167.50 (n/a)137.10 (n/a)139.70 (n/a)107.40 (n/a)20.99 (n/a)

mem_copy_4_cores_2_chans_2048_tile_512_False0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.08 (-18.19%)0.06 (-38.26%)0.06 (-41.83%)0.04 (-33.98%)0.02 (-4.28%)201.70 (+51.43%)153.98 (+65.61%)143.30 (+71.82%)100.40 (+22.29%)39.46 (+75.43%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.09 (n/a)0.10 (n/a)0.06 (n/a)0.02 (n/a)133.20 (n/a)92.98 (n/a)83.40 (n/a)82.10 (n/a)22.49 (n/a)

mem_copy_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
003b49d — 2025-11-14 19:30:540.05 (-13.48%)0.05 (-17.79%)0.05 (-15.33%)0.03 (-25.09%)0.01 (+6.67%)267.00 (+33.50%)176.80 (+24.51%)150.00 (+18.11%)141.00 (+15.57%)47.41 (+61.41%)
0999a9a — 2025-11-13 19:24:080.06 (n/a)0.06 (n/a)0.06 (n/a)0.04 (n/a)0.01 (n/a)200.00 (n/a)142.00 (n/a)127.00 (n/a)122.00 (n/a)29.37 (n/a)

mem_copy_8_cols_2_channels_2048_tile_128

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
003b49d — 2025-11-14 19:30:540.05 (+13.38%)0.05 (+23.85%)0.05 (+37.91%)0.05 (+30.72%)0.00 (-56.85%)166.00 (-23.50%)153.40 (-20.60%)153.00 (-27.49%)142.00 (-11.80%)7.63 (-70.60%)
0999a9a — 2025-11-13 19:24:080.05 (n/a)0.04 (n/a)0.04 (n/a)0.04 (n/a)0.01 (n/a)217.00 (n/a)193.20 (n/a)211.00 (n/a)161.00 (n/a)25.96 (n/a)

mem_copy_8_cores_1_chans_2048_tile_256_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.09 (+16.31%)0.08 (+15.34%)0.08 (+14.60%)0.07 (+19.30%)0.01 (-5.41%)117.30 (-16.15%)107.80 (-13.61%)108.50 (-12.78%)94.10 (-13.99%)8.29 (-32.01%)
0a6c11c — 2025-12-04 00:39:100.07 (n/a)0.07 (n/a)0.07 (n/a)0.06 (n/a)0.01 (n/a)139.90 (n/a)124.78 (n/a)124.40 (n/a)109.40 (n/a)12.20 (n/a)

mem_copy_8_cores_1_chans_2048_tile_256_False0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.07 (-27.80%)0.05 (-37.71%)0.05 (-45.06%)0.04 (-33.48%)0.01 (-29.29%)197.40 (+50.34%)160.22 (+60.57%)161.40 (+81.96%)120.40 (+38.55%)27.33 (+44.55%)
1531781 — 2026-01-23 18:49:240.09 (n/a)0.08 (n/a)0.09 (n/a)0.06 (n/a)0.01 (n/a)131.30 (n/a)99.78 (n/a)88.70 (n/a)86.90 (n/a)18.90 (n/a)

mem_copy_8_cores_2_chans_2048_tile_256_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.09 (+12.61%)0.07 (-2.24%)0.07 (-8.85%)0.06 (+5.13%)0.01 (+35.50%)136.50 (-4.88%)120.86 (+2.96%)125.10 (+9.64%)91.60 (-11.15%)15.72 (+10.46%)
0a6c11c — 2025-12-04 00:39:100.08 (n/a)0.07 (n/a)0.07 (n/a)0.06 (n/a)0.01 (n/a)143.50 (n/a)117.38 (n/a)114.10 (n/a)103.10 (n/a)14.23 (n/a)

mem_copy_8_cores_2_chans_2048_tile_256_False0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.07 (-29.05%)0.05 (-41.73%)0.05 (-45.80%)0.04 (-31.86%)0.01 (-23.69%)212.00 (+46.81%)168.72 (+72.69%)162.60 (+84.56%)112.80 (+41.00%)39.92 (+51.53%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.09 (n/a)0.09 (n/a)0.06 (n/a)0.02 (n/a)144.40 (n/a)97.70 (n/a)88.10 (n/a)80.00 (n/a)26.35 (n/a)

mha

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.18 (+0.16%)0.18 (+0.19%)0.18 (+0.12%)0.18 (+0.32%)0.00 (-52.24%)47253.90 (-0.32%)47213.70 (-0.19%)47210.90 (-0.12%)47167.70 (-0.16%)28.71 (-52.47%)
0a6c11c — 2025-12-04 00:39:100.18 (n/a)0.18 (n/a)0.18 (n/a)0.18 (n/a)0.00 (n/a)47405.00 (n/a)47301.72 (n/a)47265.90 (n/a)47241.80 (n/a)60.40 (n/a)

mha0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.18 (-0.18%)0.18 (-0.59%)0.18 (-0.73%)0.18 (-0.72%)0.00 (+99.85%)47708.70 (+0.72%)47512.04 (+0.59%)47581.20 (+0.74%)47160.10 (+0.19%)212.57 (+101.59%)
1531781 — 2026-01-23 18:49:240.18 (n/a)0.18 (n/a)0.18 (n/a)0.18 (n/a)0.00 (n/a)47367.50 (n/a)47232.68 (n/a)47232.00 (n/a)47073.00 (n/a)105.45 (n/a)

relu_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.08 (-30.39%)0.07 (-24.11%)0.08 (-23.25%)0.05 (-9.78%)0.01 (-46.06%)169.50 (+10.86%)127.84 (+25.80%)107.00 (+30.17%)104.90 (+43.70%)30.64 (-14.27%)
1531781 — 2026-01-23 18:49:240.11 (n/a)0.09 (n/a)0.10 (n/a)0.05 (n/a)0.03 (n/a)152.90 (n/a)101.62 (n/a)82.20 (n/a)73.00 (n/a)35.75 (n/a)

relu_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.10 (-3.51%)0.08 (-11.09%)0.07 (-18.14%)0.05 (+0.23%)0.01 (-16.20%)150.20 (-0.27%)112.34 (+10.94%)111.60 (+22.23%)86.10 (+3.73%)23.84 (-14.90%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.08 (n/a)0.09 (n/a)0.05 (n/a)0.02 (n/a)150.60 (n/a)101.26 (n/a)91.30 (n/a)83.00 (n/a)28.02 (n/a)

relu_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.08 (-23.42%)0.06 (-27.04%)0.07 (-31.47%)0.05 (-12.74%)0.01 (-29.63%)155.70 (+14.65%)130.66 (+35.82%)125.90 (+45.89%)108.50 (+30.57%)23.41 (+4.21%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.09 (n/a)0.09 (n/a)0.06 (n/a)0.02 (n/a)135.80 (n/a)96.20 (n/a)86.30 (n/a)83.10 (n/a)22.46 (n/a)

relu_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.07 (-32.93%)0.06 (-37.70%)0.05 (-49.44%)0.05 (-9.80%)0.01 (-51.13%)171.20 (+10.88%)143.02 (+53.26%)155.00 (+97.70%)113.00 (+49.08%)26.47 (-22.70%)
1531781 — 2026-01-23 18:49:240.11 (n/a)0.09 (n/a)0.10 (n/a)0.05 (n/a)0.02 (n/a)154.40 (n/a)93.32 (n/a)78.40 (n/a)75.80 (n/a)34.24 (n/a)

rms_norm_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.06 (-30.21%)0.06 (-29.16%)0.06 (-34.43%)0.06 (-10.88%)0.00 (-75.98%)147.60 (+12.24%)139.62 (+38.65%)140.90 (+52.49%)130.20 (+43.23%)6.58 (-62.05%)
1531781 — 2026-01-23 18:49:240.09 (n/a)0.08 (n/a)0.09 (n/a)0.06 (n/a)0.01 (n/a)131.50 (n/a)100.70 (n/a)92.40 (n/a)90.90 (n/a)17.35 (n/a)

rms_norm_1_cols_2_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.06 (-50.60%)0.05 (-50.39%)0.05 (-56.14%)0.04 (-25.82%)0.01 (-76.19%)195.70 (+34.78%)163.12 (+88.49%)164.30 (+127.88%)142.80 (+102.55%)20.90 (-36.30%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.10 (n/a)0.11 (n/a)0.06 (n/a)0.03 (n/a)145.20 (n/a)86.54 (n/a)72.10 (n/a)70.50 (n/a)32.80 (n/a)

rms_norm_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.06 (-32.38%)0.05 (-38.50%)0.05 (-45.71%)0.04 (-33.30%)0.01 (-28.91%)194.30 (+49.92%)159.94 (+62.84%)167.40 (+84.16%)132.80 (+47.88%)26.51 (+51.02%)
1531781 — 2026-01-23 18:49:240.09 (n/a)0.09 (n/a)0.09 (n/a)0.06 (n/a)0.01 (n/a)129.60 (n/a)98.22 (n/a)90.90 (n/a)89.80 (n/a)17.55 (n/a)

rms_norm_2_cols_2_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.05 (-58.53%)0.05 (-56.92%)0.05 (-54.36%)0.03 (-60.84%)0.01 (-49.61%)238.70 (+155.29%)174.28 (+134.50%)160.20 (+119.15%)150.40 (+141.03%)36.33 (+213.09%)
1531781 — 2026-01-23 18:49:240.13 (n/a)0.11 (n/a)0.11 (n/a)0.09 (n/a)0.02 (n/a)93.50 (n/a)74.32 (n/a)73.10 (n/a)62.40 (n/a)11.60 (n/a)

rms_norm_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.07 (-28.42%)0.06 (-34.77%)0.05 (-40.87%)0.05 (-26.88%)0.01 (-34.22%)178.80 (+36.80%)148.74 (+52.59%)152.60 (+69.18%)120.20 (+39.77%)22.63 (+21.23%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.09 (n/a)0.09 (n/a)0.06 (n/a)0.01 (n/a)130.70 (n/a)97.48 (n/a)90.20 (n/a)86.00 (n/a)18.66 (n/a)

rms_norm_4_cols_2_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.06 (-42.93%)0.05 (-51.49%)0.05 (-55.28%)0.04 (-46.22%)0.01 (-48.72%)233.40 (+85.98%)171.76 (+104.72%)164.30 (+123.54%)127.80 (+75.31%)39.27 (+68.85%)
1531781 — 2026-01-23 18:49:240.11 (n/a)0.10 (n/a)0.11 (n/a)0.07 (n/a)0.02 (n/a)125.50 (n/a)83.90 (n/a)73.50 (n/a)72.90 (n/a)23.26 (n/a)

rms_norm_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.06 (-34.91%)0.05 (-38.43%)0.05 (-43.91%)0.04 (-18.07%)0.01 (-63.82%)188.40 (+22.10%)168.32 (+57.81%)169.00 (+78.27%)144.50 (+53.72%)18.03 (-32.30%)
1531781 — 2026-01-23 18:49:240.09 (n/a)0.08 (n/a)0.09 (n/a)0.05 (n/a)0.01 (n/a)154.30 (n/a)106.66 (n/a)94.80 (n/a)94.00 (n/a)26.64 (n/a)

rms_norm_8_cols_2_channels_2048_tile_128

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.04 (-55.39%)0.03 (-58.91%)0.04 (-58.38%)0.02 (-59.49%)0.01 (-47.00%)413.90 (+146.81%)273.56 (+150.15%)220.10 (+140.28%)199.30 (+124.18%)94.96 (+183.29%)
1531781 — 2026-01-23 18:49:240.09 (n/a)0.08 (n/a)0.09 (n/a)0.05 (n/a)0.02 (n/a)167.70 (n/a)109.36 (n/a)91.60 (n/a)88.90 (n/a)33.52 (n/a)

rope_1_cols_2_channels_4096_tile_4096_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1b2d230 — 2026-01-13 22:38:030.19 (-21.66%)0.17 (-8.53%)0.18 (+2.09%)0.14 (-6.71%)0.02 (-38.07%)174.10 (+7.20%)150.06 (+8.11%)139.60 (-2.04%)129.80 (+27.63%)20.36 (-10.65%)
d48746f — 2026-01-12 22:17:100.24 (n/a)0.18 (n/a)0.17 (n/a)0.15 (n/a)0.04 (n/a)162.40 (n/a)138.80 (n/a)142.50 (n/a)101.70 (n/a)22.79 (n/a)

rope_1c_32rows_512cols_32arows_0m

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.76 (-30.79%)0.61 (-38.88%)0.64 (-39.58%)0.42 (-42.92%)0.13 (-12.31%)234.40 (+75.19%)169.58 (+67.01%)154.50 (+65.59%)128.70 (+44.44%)41.41 (+122.64%)
1531781 — 2026-01-23 18:49:241.10 (n/a)0.99 (n/a)1.05 (n/a)0.73 (n/a)0.15 (n/a)133.80 (n/a)101.54 (n/a)93.30 (n/a)89.10 (n/a)18.60 (n/a)

rope_1c_32rows_512cols_8arows_0m

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.57 (+3.28%)0.45 (-14.08%)0.44 (-15.42%)0.31 (-35.53%)0.12 (+303.29%)238.40 (+55.11%)174.04 (+23.57%)165.70 (+18.27%)128.30 (-3.17%)49.04 (+483.86%)
1531781 — 2026-01-23 18:49:240.56 (n/a)0.52 (n/a)0.53 (n/a)0.48 (n/a)0.03 (n/a)153.70 (n/a)140.84 (n/a)140.10 (n/a)132.50 (n/a)8.40 (n/a)

rope_2_cols_2_channels_4096_tile_2048_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1b2d230 — 2026-01-13 22:38:030.21 (-0.56%)0.17 (-10.62%)0.16 (-13.58%)0.13 (-13.33%)0.03 (+17.05%)183.70 (+15.39%)151.76 (+12.77%)155.20 (+15.73%)117.30 (+0.60%)24.24 (+35.73%)
d48746f — 2026-01-12 22:17:100.21 (n/a)0.19 (n/a)0.18 (n/a)0.15 (n/a)0.02 (n/a)159.20 (n/a)134.58 (n/a)134.10 (n/a)116.60 (n/a)17.86 (n/a)

rope_2c_32rows_512cols_32arows_0m

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.91 (-26.80%)0.64 (-34.23%)0.57 (-37.16%)0.49 (-22.18%)0.16 (-34.15%)201.70 (+28.47%)161.90 (+49.77%)172.40 (+59.19%)107.50 (+36.59%)34.81 (+11.97%)
1531781 — 2026-01-23 18:49:241.25 (n/a)0.97 (n/a)0.91 (n/a)0.63 (n/a)0.25 (n/a)157.00 (n/a)108.10 (n/a)108.30 (n/a)78.70 (n/a)31.09 (n/a)

rope_2c_32rows_512cols_8arows_0m

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.65 (+3.55%)0.43 (-24.40%)0.40 (-29.45%)0.31 (-37.22%)0.14 (+172.39%)239.80 (+59.23%)185.16 (+41.80%)184.80 (+41.72%)112.80 (-3.42%)53.09 (+322.50%)
1531781 — 2026-01-23 18:49:240.63 (n/a)0.57 (n/a)0.57 (n/a)0.49 (n/a)0.05 (n/a)150.60 (n/a)130.58 (n/a)130.40 (n/a)116.80 (n/a)12.57 (n/a)

rope_4_cols_2_channels_4096_tile_1024_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1b2d230 — 2026-01-13 22:38:030.22 (+18.92%)0.17 (+6.24%)0.17 (+5.57%)0.14 (-0.50%)0.03 (+90.77%)171.10 (+0.47%)147.40 (-4.39%)148.10 (-5.31%)110.10 (-15.89%)23.59 (+60.67%)
d48746f — 2026-01-12 22:17:100.19 (n/a)0.16 (n/a)0.16 (n/a)0.14 (n/a)0.02 (n/a)170.30 (n/a)154.16 (n/a)156.40 (n/a)130.90 (n/a)14.69 (n/a)

rope_8_cols_2_channels_4096_tile_512_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1b2d230 — 2026-01-13 22:38:030.17 (-1.30%)0.15 (+6.63%)0.15 (-8.78%)0.14 (+39.68%)0.01 (-67.35%)174.90 (-28.41%)163.42 (-10.36%)167.40 (+9.63%)146.50 (+1.31%)11.19 (-76.17%)
d48746f — 2026-01-12 22:17:100.17 (n/a)0.14 (n/a)0.16 (n/a)0.10 (n/a)0.03 (n/a)244.30 (n/a)182.30 (n/a)152.70 (n/a)144.60 (n/a)46.98 (n/a)

rope_8c_32rows_512cols_32arows_0m

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.58 (-56.42%)0.53 (-41.26%)0.52 (-38.60%)0.48 (-20.99%)0.04 (-85.33%)206.90 (+26.54%)187.96 (+60.13%)188.10 (+62.86%)168.80 (+129.35%)14.04 (-56.23%)
1531781 — 2026-01-23 18:49:241.34 (n/a)0.89 (n/a)0.85 (n/a)0.60 (n/a)0.27 (n/a)163.50 (n/a)117.38 (n/a)115.50 (n/a)73.60 (n/a)32.08 (n/a)

rope_8c_32rows_512cols_8arows_0m

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.67 (+23.28%)0.50 (-3.75%)0.39 (-26.18%)0.38 (-20.44%)0.15 (+471.69%)192.20 (+25.70%)159.22 (+11.09%)189.30 (+35.50%)109.90 (-18.89%)43.64 (+484.56%)
1531781 — 2026-01-23 18:49:240.54 (n/a)0.52 (n/a)0.53 (n/a)0.48 (n/a)0.03 (n/a)152.90 (n/a)143.32 (n/a)139.70 (n/a)135.50 (n/a)7.46 (n/a)

sigmoid_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.07 (-42.57%)0.06 (-38.64%)0.06 (-46.15%)0.05 (-5.77%)0.01 (-75.07%)162.60 (+6.14%)135.80 (+49.36%)127.30 (+85.57%)119.00 (+74.23%)17.62 (-52.13%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.10 (n/a)0.12 (n/a)0.05 (n/a)0.03 (n/a)153.20 (n/a)90.92 (n/a)68.60 (n/a)68.30 (n/a)36.80 (n/a)

sigmoid_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.08 (-37.06%)0.06 (-39.99%)0.07 (-43.19%)0.04 (-36.61%)0.02 (-40.48%)222.20 (+57.81%)148.78 (+65.61%)121.40 (+76.20%)108.20 (+58.88%)49.42 (+54.22%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.10 (n/a)0.12 (n/a)0.06 (n/a)0.03 (n/a)140.80 (n/a)89.84 (n/a)68.90 (n/a)68.10 (n/a)32.04 (n/a)

sigmoid_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.07 (-25.67%)0.06 (-27.29%)0.06 (-41.13%)0.05 (+10.69%)0.01 (-59.60%)153.10 (-9.62%)138.60 (+30.04%)148.50 (+69.91%)113.00 (+34.52%)18.08 (-50.28%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.08 (n/a)0.09 (n/a)0.05 (n/a)0.02 (n/a)169.40 (n/a)106.58 (n/a)87.40 (n/a)84.00 (n/a)36.37 (n/a)

sigmoid_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.05 (-57.84%)0.04 (-55.83%)0.04 (-63.33%)0.04 (-19.44%)0.00 (-85.83%)195.30 (+24.16%)183.72 (+104.72%)187.40 (+172.78%)156.40 (+137.33%)15.90 (-59.06%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.10 (n/a)0.12 (n/a)0.05 (n/a)0.03 (n/a)157.30 (n/a)89.74 (n/a)68.70 (n/a)65.90 (n/a)38.85 (n/a)

silu_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.07 (-30.74%)0.06 (-28.56%)0.07 (-34.67%)0.04 (-23.52%)0.01 (-49.42%)187.40 (+30.77%)141.66 (+35.07%)125.90 (+53.16%)116.30 (+44.29%)30.78 (-4.51%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.08 (n/a)0.10 (n/a)0.06 (n/a)0.02 (n/a)143.30 (n/a)104.88 (n/a)82.20 (n/a)80.60 (n/a)32.24 (n/a)

silu_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.07 (-28.88%)0.05 (-37.13%)0.05 (-46.12%)0.04 (-36.04%)0.01 (-39.86%)217.20 (+56.37%)162.14 (+57.26%)154.50 (+85.70%)114.70 (+40.56%)38.10 (+34.60%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.08 (n/a)0.10 (n/a)0.06 (n/a)0.02 (n/a)138.90 (n/a)103.10 (n/a)83.20 (n/a)81.60 (n/a)28.30 (n/a)

silu_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.07 (-41.59%)0.05 (-49.79%)0.05 (-58.44%)0.04 (-31.90%)0.01 (-55.33%)201.70 (+46.80%)159.84 (+91.75%)168.30 (+140.43%)116.60 (+71.22%)32.91 (+8.70%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.11 (n/a)0.12 (n/a)0.06 (n/a)0.03 (n/a)137.40 (n/a)83.36 (n/a)70.00 (n/a)68.10 (n/a)30.27 (n/a)

silu_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.06 (-48.14%)0.04 (-52.80%)0.05 (-44.45%)0.02 (-70.12%)0.02 (-0.33%)367.40 (+234.61%)210.24 (+135.70%)158.30 (+80.09%)137.10 (+92.83%)94.89 (+554.20%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.09 (n/a)0.09 (n/a)0.07 (n/a)0.02 (n/a)109.80 (n/a)89.20 (n/a)87.90 (n/a)71.10 (n/a)14.50 (n/a)

softmax_1_cols_2_channels_4096_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.21 (+20.19%)0.16 (+13.84%)0.16 (+8.19%)0.12 (+26.61%)0.04 (+17.18%)135.50 (-20.99%)104.68 (-12.65%)102.70 (-7.56%)76.50 (-16.85%)22.55 (-26.21%)
0f0da0b — 2026-01-14 16:31:000.18 (n/a)0.14 (n/a)0.15 (n/a)0.10 (n/a)0.03 (n/a)171.50 (n/a)119.84 (n/a)111.10 (n/a)92.00 (n/a)30.56 (n/a)

softmax_2_cols_2_channels_32768_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:511.19 (n/a)0.88 (n/a)0.91 (n/a)0.51 (n/a)0.26 (n/a)257.50 (n/a)161.36 (n/a)144.70 (n/a)110.10 (n/a)57.91 (n/a)

softmax_2_cols_2_channels_32768_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:511.05 (n/a)0.79 (n/a)0.72 (n/a)0.66 (n/a)0.15 (n/a)199.30 (n/a)170.88 (n/a)181.40 (n/a)124.90 (n/a)28.64 (n/a)

softmax_2_cols_2_channels_4096_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.22 (+40.26%)0.18 (+43.33%)0.20 (+64.08%)0.13 (+24.34%)0.04 (+68.98%)127.80 (-19.62%)94.74 (-28.95%)83.40 (-39.08%)73.90 (-28.67%)24.46 (-5.31%)
0f0da0b — 2026-01-14 16:31:000.16 (n/a)0.13 (n/a)0.12 (n/a)0.10 (n/a)0.03 (n/a)159.00 (n/a)133.34 (n/a)136.90 (n/a)103.60 (n/a)25.83 (n/a)

softmax_2_cols_2_channels_4096_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.24 (+85.21%)0.17 (+69.97%)0.15 (+54.16%)0.11 (+37.21%)0.05 (+203.82%)148.10 (-27.12%)102.78 (-37.88%)106.70 (-35.10%)69.60 (-45.96%)31.81 (+15.98%)
0f0da0b — 2026-01-14 16:31:000.13 (n/a)0.10 (n/a)0.10 (n/a)0.08 (n/a)0.02 (n/a)203.20 (n/a)165.46 (n/a)164.40 (n/a)128.80 (n/a)27.43 (n/a)

softmax_4_cols_4_channels_32768_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.60 (n/a)0.50 (n/a)0.58 (n/a)0.27 (n/a)0.14 (n/a)483.80 (n/a)289.12 (n/a)225.80 (n/a)217.90 (n/a)113.83 (n/a)

swiglu

No metrics available.

swiglu_decode_1x2048x2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.00 (-15.46%)0.00 (-27.20%)0.00 (-28.17%)0.00 (-33.90%)0.00 (+1834.83%)5187.20 (+51.28%)4720.84 (+38.38%)4750.90 (+39.22%)4015.10 (+18.29%)390.23 (+3298.45%)
0a6c11c — 2025-12-04 00:39:100.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)3428.80 (n/a)3411.62 (n/a)3412.60 (n/a)3394.20 (n/a)11.48 (n/a)

swiglu_decode_1x2048x2048_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:51n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)
1531781 — 2026-01-23 18:49:240.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)5111.40 (n/a)4324.88 (n/a)4736.10 (n/a)3410.70 (n/a)843.97 (n/a)

swiglu_prefill_256x2048x2048_0

No metrics available.

tanh_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.08 (-21.28%)0.06 (-18.76%)0.07 (-3.53%)0.05 (-16.80%)0.01 (-30.62%)169.30 (+20.24%)133.28 (+22.12%)120.20 (+3.71%)108.70 (+26.99%)26.33 (+11.97%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.08 (n/a)0.07 (n/a)0.06 (n/a)0.02 (n/a)140.80 (n/a)109.14 (n/a)115.90 (n/a)85.60 (n/a)23.52 (n/a)

tanh_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.08 (-28.14%)0.06 (-39.84%)0.06 (-37.17%)0.04 (-51.51%)0.01 (+42.75%)213.10 (+106.29%)152.60 (+73.37%)139.70 (+59.11%)108.20 (+39.07%)39.75 (+312.53%)
1531781 — 2026-01-23 18:49:240.11 (n/a)0.09 (n/a)0.09 (n/a)0.08 (n/a)0.01 (n/a)103.30 (n/a)88.02 (n/a)87.80 (n/a)77.80 (n/a)9.64 (n/a)

tanh_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.07 (-30.40%)0.06 (-38.19%)0.06 (-40.75%)0.04 (-33.63%)0.01 (-32.26%)182.90 (+50.66%)151.08 (+61.76%)145.70 (+68.83%)121.60 (+43.74%)22.95 (+45.23%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.09 (n/a)0.09 (n/a)0.07 (n/a)0.01 (n/a)121.40 (n/a)93.40 (n/a)86.30 (n/a)84.60 (n/a)15.80 (n/a)

tanh_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.07 (-38.28%)0.05 (-43.12%)0.05 (-50.79%)0.04 (-26.61%)0.01 (-54.20%)223.20 (+36.26%)171.38 (+66.55%)163.30 (+103.11%)116.10 (+61.92%)41.21 (+3.06%)
1531781 — 2026-01-23 18:49:240.11 (n/a)0.09 (n/a)0.10 (n/a)0.05 (n/a)0.03 (n/a)163.80 (n/a)102.90 (n/a)80.40 (n/a)71.70 (n/a)39.99 (n/a)

transpose_2048_M_64_N_1_cols_1_channels_64_m_64_n_8_s

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:493.93 (-7.02%)3.68 (+2.48%)3.66 (+1.82%)3.45 (+15.62%)0.15 (-63.64%)151.80 (-13.50%)142.70 (-3.59%)143.40 (-1.78%)133.50 (+7.49%)5.90 (-66.22%)
0a6c11c — 2025-12-04 00:39:104.22 (n/a)3.59 (n/a)3.59 (n/a)2.99 (n/a)0.42 (n/a)175.50 (n/a)148.02 (n/a)146.00 (n/a)124.20 (n/a)17.46 (n/a)

transpose_2048_M_64_N_1_cols_1_channels_64_m_64_n_8_s0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:513.73 (-24.53%)3.36 (-22.65%)3.39 (-22.41%)3.03 (-18.81%)0.28 (-44.59%)172.90 (+23.24%)156.88 (+28.61%)154.50 (+28.86%)140.40 (+32.45%)12.87 (-9.06%)
1531781 — 2026-01-23 18:49:244.95 (n/a)4.34 (n/a)4.37 (n/a)3.74 (n/a)0.50 (n/a)140.30 (n/a)121.98 (n/a)119.90 (n/a)106.00 (n/a)14.15 (n/a)

transpose_2048_M_64_N_1_cols_2_channels_64_m_64_n_8_s

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:495.11 (+15.57%)3.85 (-3.32%)3.40 (-18.64%)3.29 (+7.09%)0.71 (+49.14%)159.60 (-6.61%)140.18 (+4.80%)154.20 (+22.97%)102.60 (-13.49%)22.70 (+19.77%)
0a6c11c — 2025-12-04 00:39:104.42 (n/a)3.99 (n/a)4.18 (n/a)3.07 (n/a)0.48 (n/a)170.90 (n/a)133.76 (n/a)125.40 (n/a)118.60 (n/a)18.95 (n/a)

transpose_2048_M_64_N_1_cols_2_channels_64_m_64_n_8_s0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:514.66 (-23.30%)3.44 (-31.13%)3.27 (-35.68%)2.78 (-32.10%)0.76 (+2.43%)188.60 (+47.23%)157.88 (+47.74%)160.20 (+55.53%)112.50 (+30.36%)30.86 (+96.96%)
1531781 — 2026-01-23 18:49:246.07 (n/a)4.99 (n/a)5.09 (n/a)4.09 (n/a)0.74 (n/a)128.10 (n/a)106.86 (n/a)103.00 (n/a)86.30 (n/a)15.67 (n/a)

weighted_rms_norm_1_cols_2_channels_2048_weights_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.09 (-18.03%)0.07 (-14.31%)0.08 (-16.96%)0.05 (-4.57%)0.01 (-27.38%)243.70 (+57.23%)171.38 (+71.52%)161.90 (+80.69%)140.00 (+83.01%)42.02 (+33.98%)
1531781 — 2026-01-23 18:49:240.11 (n/a)0.09 (n/a)0.09 (n/a)0.05 (n/a)0.02 (n/a)155.00 (n/a)99.92 (n/a)89.60 (n/a)76.50 (n/a)31.36 (n/a)

weighted_rms_norm_2_cols_2_channels_2048_weights_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.08 (-23.16%)0.07 (-21.02%)0.07 (-29.19%)0.06 (+26.08%)0.01 (-74.50%)165.20 (-0.90%)152.36 (+48.33%)156.20 (+76.50%)135.20 (+62.70%)11.30 (-68.48%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.09 (n/a)0.09 (n/a)0.05 (n/a)0.02 (n/a)166.70 (n/a)102.72 (n/a)88.50 (n/a)83.10 (n/a)35.87 (n/a)

weighted_rms_norm_4_cols_2_channels_2048_weights_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.06 (-30.61%)0.05 (-39.02%)0.05 (-41.03%)0.04 (-41.32%)0.01 (-9.17%)228.20 (+91.76%)180.20 (+86.77%)171.40 (+90.66%)144.70 (+62.04%)32.36 (+152.71%)
1531781 — 2026-01-23 18:49:240.09 (n/a)0.09 (n/a)0.09 (n/a)0.07 (n/a)0.01 (n/a)119.00 (n/a)96.48 (n/a)89.90 (n/a)89.30 (n/a)12.81 (n/a)

weighted_rms_norm_8_cols_2_channels_2048_weights_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
71486ae — 2026-02-09 15:31:510.05 (-50.92%)0.05 (-49.10%)0.05 (-50.67%)0.04 (-32.24%)0.01 (-71.35%)212.90 (+56.77%)180.86 (+101.13%)172.40 (+115.50%)160.90 (+116.55%)22.43 (-12.94%)
1531781 — 2026-01-23 18:49:240.11 (n/a)0.10 (n/a)0.10 (n/a)0.06 (n/a)0.02 (n/a)135.80 (n/a)89.92 (n/a)80.00 (n/a)74.30 (n/a)25.76 (n/a)

@github-actions
Copy link
Contributor

github-actions bot commented Feb 9, 2026

📊 Test Results for Test Example Applications

71486ae (2026_02_09_15_43_18)

IRONCLAD

Tested on 2026_02_09_15_43_18 at commit 71486ae.

Test Checks TTFT (mean)TPS (mean)
llama_3.2_1b_prompt_13_tokens_1 ❌ 0/1 n/a n/a
llama_3.2_1b_prompt_13_tokens_40 ❌ 0/1 n/a n/a
llama_3.2_1b_prompt_2048_tokens_1 ❌ 0/1 n/a n/a
llama_3.2_1b_prompt_2048_tokens_40 ❌ 0/1 n/a n/a
📈 Trends (vs main branch) for Test Example Applications

71486ae (2026_02_09_15_43_18)

IRONCLAD Trends

llama_3.2_1b

Commit/Date Num Tokens (max)Num Tokens (mean)Num Tokens (median)Num Tokens (min)Num Tokens (stddev)TPS (max)TPS (mean)TPS (median)TPS (min)TPS (stddev)TTFT (max)TTFT (mean)TTFT (median)TTFT (min)TTFT (stddev)Total (max)Total (mean)Total (median)Total (min)Total (stddev)
130b6ea — 2025-12-05 21:33:1240.00 (+0.00%)40.00 (+0.00%)40.00 (+0.00%)40.00 (+0.00%)0.00 (n/a)4.71 (-0.42%)4.64 (-0.09%)4.64 (+0.65%)4.55 (-0.22%)0.05 (-17.66%)4.41 (-0.34%)4.39 (-0.19%)4.38 (-0.33%)4.37 (-0.15%)0.01 (-25.90%)12.96 (-0.00%)12.80 (+0.07%)12.80 (-0.23%)12.67 (+0.44%)0.09 (-21.12%)
0a6c11c — 2025-12-03 23:35:1540.00 (n/a)40.00 (n/a)40.00 (n/a)40.00 (n/a)0.00 (n/a)4.73 (n/a)4.64 (n/a)4.61 (n/a)4.56 (n/a)0.06 (n/a)4.42 (n/a)4.40 (n/a)4.40 (n/a)4.37 (n/a)0.02 (n/a)12.96 (n/a)12.79 (n/a)12.83 (n/a)12.62 (n/a)0.12 (n/a)

llama_3.2_1b_prompt_13_tokens_1

Commit/Date Num_Tokens (max)Num_Tokens (mean)Num_Tokens (median)Num_Tokens (min)Num_Tokens (stddev)TPS (max)TPS (mean)TPS (median)TPS (min)TPS (stddev)TTFT (max)TTFT (mean)TTFT (median)TTFT (min)TTFT (stddev)
71486ae — 2026-02-09 15:42:47n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)
1531781 — 2026-01-23 18:40:341.00 (n/a)1.00 (n/a)1.00 (n/a)1.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.61 (n/a)0.61 (n/a)0.61 (n/a)0.61 (n/a)0.00 (n/a)

llama_3.2_1b_prompt_13_tokens_40

Commit/Date Num_Tokens (max)Num_Tokens (mean)Num_Tokens (median)Num_Tokens (min)Num_Tokens (stddev)TPS (max)TPS (mean)TPS (median)TPS (min)TPS (stddev)TTFT (max)TTFT (mean)TTFT (median)TTFT (min)TTFT (stddev)
71486ae — 2026-02-09 15:42:47n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)
1531781 — 2026-01-23 18:40:3440.00 (n/a)40.00 (n/a)40.00 (n/a)40.00 (n/a)0.00 (n/a)4.34 (n/a)4.34 (n/a)4.34 (n/a)4.34 (n/a)0.00 (n/a)0.61 (n/a)0.61 (n/a)0.61 (n/a)0.61 (n/a)0.00 (n/a)

llama_3.2_1b_prompt_2048_tokens_1

Commit/Date Num_Tokens (max)Num_Tokens (mean)Num_Tokens (median)Num_Tokens (min)Num_Tokens (stddev)TPS (max)TPS (mean)TPS (median)TPS (min)TPS (stddev)TTFT (max)TTFT (mean)TTFT (median)TTFT (min)TTFT (stddev)
71486ae — 2026-02-09 15:42:47n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)
1531781 — 2026-01-23 18:40:341.00 (n/a)1.00 (n/a)1.00 (n/a)1.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)2.70 (n/a)2.70 (n/a)2.70 (n/a)2.70 (n/a)0.00 (n/a)

llama_3.2_1b_prompt_2048_tokens_40

Commit/Date Num_Tokens (max)Num_Tokens (mean)Num_Tokens (median)Num_Tokens (min)Num_Tokens (stddev)TPS (max)TPS (mean)TPS (median)TPS (min)TPS (stddev)TTFT (max)TTFT (mean)TTFT (median)TTFT (min)TTFT (stddev)
71486ae — 2026-02-09 15:42:47n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)
1531781 — 2026-01-23 18:40:3440.00 (n/a)40.00 (n/a)40.00 (n/a)40.00 (n/a)0.00 (n/a)4.22 (n/a)4.22 (n/a)4.22 (n/a)4.22 (n/a)0.00 (n/a)2.70 (n/a)2.70 (n/a)2.70 (n/a)2.70 (n/a)0.00 (n/a)

@github-actions
Copy link
Contributor

github-actions bot commented Feb 9, 2026

📊 Test Results for Small Benchmark/Test Suite

9369dae (2026_02_09_20_38_58)

IRONCLAD

Tested on 2026_02_09_20_38_58 at commit 9369dae.

Test Checks Latency (mean)Bandwidth (mean)GFLOP/s (mean)
axpy_1_cols_2_channels_2048_tile_2048_3.0 ✅ 5/5 151.16 0.08 n/a
axpy_2_cols_2_channels_2048_tile_1024_3.0 ✅ 5/5 154.72 0.08 n/a
axpy_4_cols_2_channels_2048_tile_512_3.0 ✅ 5/5 190.34 0.07 n/a
axpy_8_cols_2_channels_2048_tile_256_3.0 ✅ 5/5 263.54 0.05 n/a
dequant_1_cols_1_channels_2048_tile_2048 ✅ 5/5 158.00 0.03 n/a
dequant_1_cols_2_channels_2048_tile_1024 ✅ 5/5 129.84 0.04 n/a
dequant_2_cols_1_channels_2048_tile_1024 ✅ 5/5 143.34 0.04 n/a
dequant_2_cols_2_channels_2048_tile_512 ✅ 5/5 154.16 0.04 n/a
dequant_4_cols_1_channels_2048_tile_512 ✅ 5/5 126.74 0.04 n/a
dequant_4_cols_2_channels_2048_tile_256 ✅ 5/5 157.22 0.04 n/a
dequant_8_cols_1_channels_2048_tile_256 ✅ 5/5 169.72 0.03 n/a
dequant_8_cols_2_channels_2048_tile_128 ✅ 5/5 213.02 0.03 n/a
eltwise_add_1_cols_2_channels_2048_tile_2048 ✅ 5/5 123.22 0.10 n/a
eltwise_add_2_cols_2_channels_2048_tile_1024 ✅ 5/5 152.08 0.08 n/a
eltwise_add_4_cols_2_channels_2048_tile_512 ✅ 5/5 160.52 0.08 n/a
eltwise_add_8_cols_2_channels_2048_tile_256 ✅ 5/5 166.70 0.07 n/a
eltwise_mul_1_cols_2_channels_2048_tile_2048 ✅ 5/5 132.46 0.10 n/a
eltwise_mul_2_cols_2_channels_2048_tile_1024 ✅ 5/5 140.62 0.09 n/a
eltwise_mul_4_cols_2_channels_2048_tile_512 ✅ 5/5 134.76 0.10 n/a
eltwise_mul_8_cols_2_channels_2048_tile_256 ✅ 5/5 218.40 0.06 n/a
gelu_1_cols_1_channels_2048_tile_2048 ✅ 5/5 167.90 0.05 n/a
gelu_1_cols_2_channels_2048_tile_1024 ✅ 5/5 151.52 0.06 n/a
gelu_2_cols_1_channels_2048_tile_1024 ✅ 5/5 155.80 0.06 n/a
gelu_2_cols_2_channels_2048_tile_512 ✅ 5/5 146.68 0.06 n/a
gelu_4_cols_1_channels_2048_tile_512 ✅ 5/5 178.68 0.05 n/a
gelu_4_cols_2_channels_2048_tile_256 ✅ 5/5 163.96 0.05 n/a
gelu_8_cols_1_channels_2048_tile_256 ✅ 5/5 185.08 0.04 n/a
gelu_8_cols_2_channels_2048_tile_128 ✅ 5/5 199.86 0.04 n/a
gemm_1792x896x1152_64x32x48_8cols_ccolmaj ✅ 5/5 2199.56 4.31 n/a
gemm_192x384x64_48x96x16_4cols ✅ 5/5 270.76 0.87 n/a
gemm_192x384x64_48x96x16_4cols_bcolmaj_ccolmaj ✅ 5/5 234.42 1.04 n/a
gemm_2048x2048x2048_64x64x64_1cols ✅ 5/5 48617.40 0.52 n/a
gemm_2048x2048x2048_64x64x64_2cols_bcolmaj ✅ 5/5 27767.00 0.91 n/a
gemm_2048x2048x2048_64x64x64_8cols_bcolmaj_ccolmaj ✅ 5/5 7655.12 3.29 n/a
gemm_384x1536x1792_32x48x64_4cols_bcolmaj ✅ 5/5 2090.82 3.87 n/a
gemm_896x1792x640_32x64x80_8cols_ccolmaj ✅ 5/5 1314.06 5.28 n/a
layer_norm_1_cols_1_channels_2048_tile_2048 ✅ 5/5 152.78 0.05 n/a
layer_norm_1_cols_2_channels_2048_tile_1024 ✅ 5/5 165.40 0.05 n/a
layer_norm_2_cols_1_channels_2048_tile_1024 ✅ 5/5 166.40 0.05 n/a
layer_norm_2_cols_2_channels_2048_tile_512 ✅ 5/5 167.08 0.05 n/a
layer_norm_4_cols_1_channels_2048_tile_512 ✅ 5/5 217.24 0.04 n/a
layer_norm_4_cols_2_channels_2048_tile_256 ✅ 5/5 224.72 0.04 n/a
layer_norm_8_cols_1_channels_2048_tile_256 ✅ 5/5 177.72 0.05 n/a
layer_norm_8_cols_2_channels_2048_tile_128 ✅ 5/5 212.44 0.04 n/a
matrix_vector_mul_128x128_32tsi_128tso_1col ✅ 5/5 n/a 0.22 n/a
matrix_vector_mul_2048x8192_1tsi_1024tso_2col ✅ 5/5 n/a 24.11 n/a
matrix_vector_mul_2048x8192_1tsi_2048tso_1col ✅ 5/5 n/a 12.46 n/a
matrix_vector_mul_2048x8192_1tsi_256tso_8col ✅ 5/5 n/a 43.29 n/a
matrix_vector_mul_2048x8192_1tsi_512tso_4col ✅ 5/5 n/a 39.41 n/a
matrix_vector_mul_8192x2048_4tsi_1024tso_1col ✅ 5/5 n/a 13.26 n/a
matrix_vector_mul_8192x2048_4tsi_1024tso_2col ✅ 5/5 n/a 24.32 n/a
matrix_vector_mul_8192x2048_4tsi_1024tso_4col ✅ 5/5 n/a 39.89 n/a
matrix_vector_mul_8192x2048_4tsi_1024tso_8col ✅ 5/5 n/a 42.95 n/a
mem_copy_16_cores_2_chans_2048_tile_128_False ✅ 5/5 171.18 0.05 n/a
mem_copy_1_cores_1_chans_2048_tile_2048_False ✅ 5/5 142.80 0.06 n/a
mem_copy_2_cores_1_chans_2048_tile_1024_False ✅ 5/5 133.28 0.06 n/a
mem_copy_2_cores_2_chans_2048_tile_1024_False ✅ 5/5 127.84 0.07 n/a
mem_copy_4_cores_1_chans_2048_tile_512_False ✅ 5/5 144.10 0.06 n/a
mem_copy_4_cores_2_chans_2048_tile_512_False ✅ 5/5 139.24 0.06 n/a
mem_copy_8_cores_1_chans_2048_tile_256_False ✅ 5/5 166.22 0.05 n/a
mem_copy_8_cores_2_chans_2048_tile_256_False ✅ 5/5 160.92 0.05 n/a
mha ✅ 5/5 47575.70 0.18 n/a
relu_1_cols_1_channels_2048_tile_2048 ✅ 5/5 164.32 0.05 n/a
relu_2_cols_1_channels_2048_tile_1024 ✅ 5/5 151.64 0.05 n/a
relu_4_cols_1_channels_2048_tile_512 ✅ 5/5 150.52 0.06 n/a
relu_8_cols_1_channels_2048_tile_256 ✅ 5/5 173.06 0.05 n/a
rms_norm_1_cols_1_channels_2048_tile_2048 ✅ 5/5 152.10 0.05 n/a
rms_norm_1_cols_2_channels_2048_tile_1024 ✅ 5/5 149.98 0.05 n/a
rms_norm_2_cols_1_channels_2048_tile_1024 ✅ 5/5 154.28 0.05 n/a
rms_norm_2_cols_2_channels_2048_tile_512 ✅ 5/5 207.82 0.04 n/a
rms_norm_4_cols_1_channels_2048_tile_512 ✅ 5/5 183.44 0.05 n/a
rms_norm_4_cols_2_channels_2048_tile_256 ✅ 5/5 176.18 0.05 n/a
rms_norm_8_cols_1_channels_2048_tile_256 ✅ 5/5 175.68 0.05 n/a
rms_norm_8_cols_2_channels_2048_tile_128 ✅ 5/5 203.90 0.04 n/a
rope_1c_32rows_512cols_32arows_0m ✅ 5/5 159.24 0.65 n/a
rope_1c_32rows_512cols_8arows_0m ✅ 5/5 156.32 0.49 n/a
rope_2c_32rows_512cols_32arows_0m ✅ 5/5 164.54 0.62 n/a
rope_2c_32rows_512cols_8arows_0m ✅ 5/5 173.04 0.45 n/a
rope_8c_32rows_512cols_32arows_0m ✅ 5/5 181.76 0.57 n/a
rope_8c_32rows_512cols_8arows_0m ✅ 5/5 178.62 0.44 n/a
sigmoid_1_cols_1_channels_2048_tile_2048 ✅ 5/5 138.18 0.06 n/a
sigmoid_2_cols_1_channels_2048_tile_1024 ✅ 5/5 148.86 0.06 n/a
sigmoid_4_cols_1_channels_2048_tile_512 ✅ 5/5 170.38 0.05 n/a
sigmoid_8_cols_1_channels_2048_tile_256 ✅ 5/5 182.16 0.05 n/a
silu_1_cols_1_channels_2048_tile_2048 ✅ 5/5 172.38 0.05 n/a
silu_2_cols_1_channels_2048_tile_1024 ✅ 5/5 187.30 0.05 n/a
silu_4_cols_1_channels_2048_tile_512 ✅ 5/5 157.60 0.06 n/a
silu_8_cols_1_channels_2048_tile_256 ✅ 5/5 190.12 0.04 n/a
softmax_2_cols_2_channels_32768_tile_1024 ✅ 5/5 178.12 0.77 n/a
softmax_2_cols_2_channels_32768_tile_512 ✅ 5/5 154.70 0.89 n/a
softmax_4_cols_4_channels_32768_tile_2048 ✅ 5/5 223.14 0.59 n/a
swiglu_decode_1x2048x2048 ✅ 5/5 n/a n/a n/a
swiglu_prefill_256x2048x2048 ❌ 0/5 n/a n/a n/a
tanh_1_cols_1_channels_2048_tile_2048 ✅ 5/5 172.58 0.05 n/a
tanh_2_cols_1_channels_2048_tile_1024 ✅ 5/5 161.90 0.05 n/a
tanh_4_cols_1_channels_2048_tile_512 ✅ 5/5 165.32 0.05 n/a
tanh_8_cols_1_channels_2048_tile_256 ✅ 5/5 179.78 0.05 n/a
transpose_2048_M_64_N_1_cols_1_channels_64_m_64_n_8_s ✅ 5/5 181.42 2.97 n/a
transpose_2048_M_64_N_1_cols_2_channels_64_m_64_n_8_s ✅ 5/5 150.72 3.52 n/a
weighted_rms_norm_1_cols_2_channels_2048_weights_2048 ✅ 5/5 148.78 0.08 n/a
weighted_rms_norm_2_cols_2_channels_2048_weights_1024 ✅ 5/5 163.12 0.07 n/a
weighted_rms_norm_4_cols_2_channels_2048_weights_512 ✅ 5/5 156.76 0.06 n/a
weighted_rms_norm_8_cols_2_channels_2048_weights_256 ✅ 5/5 205.40 0.04 n/a
📈 Trends (vs main branch) for Small Benchmark/Test Suite

9369dae (2026_02_09_20_38_58)

IRONCLAD Trends

axpy_1_cols_2_channels_2048_tile_2048_3.0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.10 (-21.00%)0.08 (-19.45%)0.09 (-17.50%)0.06 (-22.93%)0.02 (-4.23%)210.80 (+29.80%)151.16 (+24.97%)141.90 (+21.28%)119.40 (+26.48%)36.26 (+56.89%)
130b6ea — 2025-12-05 21:15:490.13 (n/a)0.11 (n/a)0.10 (n/a)0.08 (n/a)0.02 (n/a)162.40 (n/a)120.96 (n/a)117.00 (n/a)94.40 (n/a)23.11 (n/a)

axpy_1_cols_2_channels_2048_tile_2048_3.0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.16 (+80.10%)0.13 (+54.74%)0.11 (+34.32%)0.10 (+39.85%)0.03 (+432.10%)118.80 (-28.48%)100.30 (-33.22%)110.40 (-25.56%)78.00 (-44.48%)20.49 (+105.98%)
0f0da0b — 2026-01-14 16:31:000.09 (n/a)0.08 (n/a)0.08 (n/a)0.07 (n/a)0.01 (n/a)166.10 (n/a)150.20 (n/a)148.30 (n/a)140.50 (n/a)9.95 (n/a)

axpy_2_cols_2_channels_2048_tile_1024_3.0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.10 (-11.01%)0.08 (-11.96%)0.08 (-3.72%)0.06 (-22.44%)0.01 (+9.98%)196.00 (+28.95%)154.72 (+14.22%)148.30 (+3.85%)129.20 (+12.35%)24.82 (+65.09%)
130b6ea — 2025-12-05 21:15:490.11 (n/a)0.09 (n/a)0.09 (n/a)0.08 (n/a)0.01 (n/a)152.00 (n/a)135.46 (n/a)142.80 (n/a)115.00 (n/a)15.03 (n/a)

axpy_2_cols_2_channels_2048_tile_1024_3.0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.14 (+54.25%)0.12 (+45.65%)0.14 (+61.31%)0.08 (-2.79%)0.03 (+557.87%)156.90 (+2.89%)104.46 (-28.09%)90.30 (-37.98%)87.90 (-35.22%)29.52 (+351.33%)
0f0da0b — 2026-01-14 16:31:000.09 (n/a)0.08 (n/a)0.08 (n/a)0.08 (n/a)0.00 (n/a)152.50 (n/a)145.26 (n/a)145.60 (n/a)135.70 (n/a)6.54 (n/a)

axpy_4_cols_2_channels_2048_tile_512_3.0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.09 (-23.92%)0.07 (-32.47%)0.06 (-38.39%)0.05 (-25.80%)0.01 (-16.56%)224.60 (+34.73%)190.34 (+47.80%)197.60 (+62.23%)136.60 (+31.47%)32.40 (+41.18%)
130b6ea — 2025-12-05 21:15:490.12 (n/a)0.10 (n/a)0.10 (n/a)0.07 (n/a)0.02 (n/a)166.70 (n/a)128.78 (n/a)121.80 (n/a)103.90 (n/a)22.95 (n/a)

axpy_4_cols_2_channels_2048_tile_512_3.0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.16 (+65.19%)0.14 (+57.55%)0.15 (+63.48%)0.09 (+24.56%)0.03 (+220.28%)134.10 (-19.70%)92.44 (-34.51%)83.40 (-38.81%)78.60 (-39.49%)23.38 (+58.01%)
0f0da0b — 2026-01-14 16:31:000.09 (n/a)0.09 (n/a)0.09 (n/a)0.07 (n/a)0.01 (n/a)167.00 (n/a)141.16 (n/a)136.30 (n/a)129.90 (n/a)14.80 (n/a)

axpy_8_cols_2_channels_2048_tile_256_3.0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.06 (-40.38%)0.05 (-39.86%)0.05 (-36.47%)0.03 (-53.63%)0.01 (-0.57%)397.70 (+115.67%)263.54 (+72.05%)226.70 (+57.32%)213.10 (+67.66%)77.17 (+262.97%)
130b6ea — 2025-12-05 21:15:490.10 (n/a)0.08 (n/a)0.09 (n/a)0.07 (n/a)0.01 (n/a)184.40 (n/a)153.18 (n/a)144.10 (n/a)127.10 (n/a)21.26 (n/a)

axpy_8_cols_2_channels_2048_tile_256_3.0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.17 (+67.03%)0.14 (+66.86%)0.15 (+90.72%)0.08 (+20.53%)0.03 (+162.97%)146.80 (-17.06%)92.24 (-37.11%)80.00 (-47.54%)72.30 (-40.15%)31.07 (+36.92%)
0f0da0b — 2026-01-14 16:31:000.10 (n/a)0.09 (n/a)0.08 (n/a)0.07 (n/a)0.01 (n/a)177.00 (n/a)146.68 (n/a)152.50 (n/a)120.80 (n/a)22.69 (n/a)

dequant_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.04 (-11.89%)0.03 (-10.70%)0.03 (-13.31%)0.03 (-12.27%)0.00 (-2.13%)182.90 (+14.03%)158.00 (+11.96%)157.40 (+15.40%)129.70 (+13.47%)21.68 (+24.79%)
130b6ea — 2025-12-05 21:15:490.05 (n/a)0.04 (n/a)0.04 (n/a)0.03 (n/a)0.00 (n/a)160.40 (n/a)141.12 (n/a)136.40 (n/a)114.30 (n/a)17.37 (n/a)

dequant_1_cols_1_channels_2048_tile_2048_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.06 (+13.11%)0.05 (+1.55%)0.04 (-0.75%)0.04 (+7.05%)0.01 (+25.85%)125.50 (-6.55%)115.08 (-1.12%)121.60 (+0.75%)88.10 (-11.63%)15.40 (+4.09%)
0f0da0b — 2026-01-14 16:31:000.05 (n/a)0.05 (n/a)0.04 (n/a)0.04 (n/a)0.01 (n/a)134.30 (n/a)116.38 (n/a)120.70 (n/a)99.70 (n/a)14.79 (n/a)

dequant_1_cols_2_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.05 (-19.51%)0.04 (-17.80%)0.05 (-15.09%)0.03 (-22.07%)0.01 (-0.44%)171.40 (+28.29%)129.84 (+22.40%)112.10 (+17.88%)109.20 (+24.23%)27.66 (+54.56%)
130b6ea — 2025-12-05 21:15:490.06 (n/a)0.05 (n/a)0.06 (n/a)0.04 (n/a)0.01 (n/a)133.60 (n/a)106.08 (n/a)95.10 (n/a)87.90 (n/a)17.90 (n/a)

dequant_1_cols_2_channels_2048_tile_1024_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.07 (+24.45%)0.05 (+9.74%)0.05 (-2.94%)0.03 (+16.88%)0.01 (+14.24%)163.70 (-14.47%)114.66 (-9.57%)111.30 (+3.06%)76.50 (-19.64%)31.63 (-21.04%)
0f0da0b — 2026-01-14 16:31:000.06 (n/a)0.04 (n/a)0.05 (n/a)0.03 (n/a)0.01 (n/a)191.40 (n/a)126.80 (n/a)108.00 (n/a)95.20 (n/a)40.06 (n/a)

dequant_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.04 (-35.10%)0.04 (-31.47%)0.04 (-33.76%)0.03 (-13.72%)0.00 (-51.16%)166.20 (+15.90%)143.34 (+41.89%)147.30 (+50.92%)121.60 (+54.12%)18.89 (-16.29%)
130b6ea — 2025-12-05 21:15:490.07 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)143.40 (n/a)101.02 (n/a)97.60 (n/a)78.90 (n/a)22.56 (n/a)

dequant_2_cols_1_channels_2048_tile_1024_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.05 (+2.33%)0.05 (+22.41%)0.05 (+32.19%)0.04 (+7.46%)0.01 (-3.28%)147.80 (-6.93%)113.14 (-18.51%)106.50 (-24.31%)101.40 (-2.31%)19.61 (-8.79%)
0f0da0b — 2026-01-14 16:31:000.05 (n/a)0.04 (n/a)0.04 (n/a)0.03 (n/a)0.01 (n/a)158.80 (n/a)138.84 (n/a)140.70 (n/a)103.80 (n/a)21.51 (n/a)

dequant_2_cols_2_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.06 (+16.65%)0.04 (-8.80%)0.04 (-17.29%)0.02 (-43.39%)0.01 (+96.85%)282.00 (+76.58%)154.16 (+22.04%)131.80 (+20.92%)91.60 (-14.23%)73.83 (+228.43%)
130b6ea — 2025-12-05 21:15:490.05 (n/a)0.04 (n/a)0.05 (n/a)0.03 (n/a)0.01 (n/a)159.70 (n/a)126.32 (n/a)109.00 (n/a)106.80 (n/a)22.48 (n/a)

dequant_2_cols_2_channels_2048_tile_512_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.05 (+14.83%)0.04 (+10.04%)0.05 (+21.96%)0.03 (-12.01%)0.01 (+143.96%)158.30 (+13.64%)121.38 (-6.49%)106.20 (-17.99%)98.20 (-12.94%)26.13 (+141.52%)
0f0da0b — 2026-01-14 16:31:000.05 (n/a)0.04 (n/a)0.04 (n/a)0.04 (n/a)0.00 (n/a)139.30 (n/a)129.80 (n/a)129.50 (n/a)112.80 (n/a)10.82 (n/a)

dequant_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.05 (-17.52%)0.04 (-5.39%)0.05 (+0.63%)0.03 (-11.66%)0.01 (-15.07%)176.70 (+13.20%)126.74 (+4.87%)114.20 (-0.61%)107.00 (+21.18%)28.77 (+17.93%)
130b6ea — 2025-12-05 21:15:490.06 (n/a)0.05 (n/a)0.05 (n/a)0.03 (n/a)0.01 (n/a)156.10 (n/a)120.86 (n/a)114.90 (n/a)88.30 (n/a)24.40 (n/a)

dequant_4_cols_1_channels_2048_tile_512_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.06 (+28.85%)0.05 (+16.79%)0.05 (+23.60%)0.03 (-12.66%)0.01 (+157.83%)164.80 (+14.52%)113.62 (-10.20%)103.20 (-19.12%)84.80 (-22.42%)32.55 (+130.11%)
0f0da0b — 2026-01-14 16:31:000.05 (n/a)0.04 (n/a)0.04 (n/a)0.04 (n/a)0.00 (n/a)143.90 (n/a)126.52 (n/a)127.60 (n/a)109.30 (n/a)14.15 (n/a)

dequant_4_cols_2_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.05 (-4.74%)0.04 (-11.21%)0.03 (-20.80%)0.02 (-22.51%)0.01 (+39.06%)217.00 (+29.09%)157.22 (+15.82%)166.90 (+26.25%)112.80 (+5.03%)41.84 (+82.66%)
130b6ea — 2025-12-05 21:15:490.05 (n/a)0.04 (n/a)0.04 (n/a)0.03 (n/a)0.01 (n/a)168.10 (n/a)135.74 (n/a)132.20 (n/a)107.40 (n/a)22.91 (n/a)

dequant_4_cols_2_channels_2048_tile_256_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.07 (+42.90%)0.05 (+60.94%)0.06 (+84.55%)0.03 (+50.15%)0.01 (+27.03%)188.40 (-33.40%)112.72 (-39.08%)93.80 (-45.84%)79.70 (-29.96%)43.58 (-36.67%)
0f0da0b — 2026-01-14 16:31:000.05 (n/a)0.03 (n/a)0.03 (n/a)0.02 (n/a)0.01 (n/a)282.90 (n/a)185.04 (n/a)173.20 (n/a)113.80 (n/a)68.82 (n/a)

dequant_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.04 (-3.50%)0.03 (-16.19%)0.03 (-18.42%)0.03 (-18.51%)0.01 (+41.72%)209.10 (+22.71%)169.72 (+21.12%)166.40 (+22.53%)124.00 (+3.68%)31.60 (+76.15%)
130b6ea — 2025-12-05 21:15:490.04 (n/a)0.04 (n/a)0.04 (n/a)0.03 (n/a)0.00 (n/a)170.40 (n/a)140.12 (n/a)135.80 (n/a)119.60 (n/a)17.94 (n/a)

dequant_8_cols_1_channels_2048_tile_256_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.04 (-5.92%)0.04 (-11.07%)0.04 (-12.74%)0.03 (-9.17%)0.00 (-4.85%)166.90 (+10.09%)138.66 (+12.51%)135.90 (+14.59%)117.60 (+6.33%)17.94 (+10.10%)
0f0da0b — 2026-01-14 16:31:000.05 (n/a)0.04 (n/a)0.04 (n/a)0.03 (n/a)0.00 (n/a)151.60 (n/a)123.24 (n/a)118.60 (n/a)110.60 (n/a)16.30 (n/a)

dequant_8_cols_2_channels_2048_tile_128

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.03 (-34.98%)0.03 (-26.78%)0.02 (-24.64%)0.02 (-15.19%)0.00 (-48.96%)233.70 (+17.91%)213.02 (+33.14%)220.50 (+32.67%)163.80 (+53.80%)28.78 (-3.61%)
130b6ea — 2025-12-05 21:15:490.05 (n/a)0.03 (n/a)0.03 (n/a)0.03 (n/a)0.01 (n/a)198.20 (n/a)160.00 (n/a)166.20 (n/a)106.50 (n/a)29.86 (n/a)

dequant_8_cols_2_channels_2048_tile_128_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.05 (+10.58%)0.04 (+11.15%)0.04 (+32.75%)0.02 (-15.30%)0.01 (+23.92%)229.90 (+18.08%)145.60 (-7.34%)131.00 (-24.67%)106.20 (-9.54%)49.03 (+42.68%)
0f0da0b — 2026-01-14 16:31:000.04 (n/a)0.03 (n/a)0.03 (n/a)0.03 (n/a)0.01 (n/a)194.70 (n/a)157.14 (n/a)173.90 (n/a)117.40 (n/a)34.37 (n/a)

eltwise_add_1_cols_2_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.11 (-32.12%)0.10 (-16.90%)0.10 (+1.88%)0.08 (+7.14%)0.01 (-68.31%)145.40 (-6.68%)123.22 (+12.24%)118.30 (-1.83%)108.70 (+47.29%)15.76 (-54.18%)
1531781 — 2026-01-23 18:49:240.17 (n/a)0.12 (n/a)0.10 (n/a)0.08 (n/a)0.04 (n/a)155.80 (n/a)109.78 (n/a)120.50 (n/a)73.80 (n/a)34.38 (n/a)

eltwise_add_2_cols_2_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.10 (-31.33%)0.08 (-34.23%)0.08 (-42.87%)0.07 (-21.12%)0.02 (-35.63%)186.00 (+26.79%)152.08 (+50.28%)160.50 (+75.03%)118.50 (+45.58%)28.81 (+10.84%)
1531781 — 2026-01-23 18:49:240.15 (n/a)0.13 (n/a)0.13 (n/a)0.08 (n/a)0.03 (n/a)146.70 (n/a)101.20 (n/a)91.70 (n/a)81.40 (n/a)26.00 (n/a)

eltwise_add_4_cols_2_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.10 (-42.99%)0.08 (-46.12%)0.07 (-50.75%)0.07 (-38.39%)0.01 (-40.15%)188.40 (+62.41%)160.52 (+85.66%)164.30 (+103.09%)127.60 (+75.52%)28.85 (+67.50%)
1531781 — 2026-01-23 18:49:240.17 (n/a)0.15 (n/a)0.15 (n/a)0.11 (n/a)0.02 (n/a)116.00 (n/a)86.46 (n/a)80.90 (n/a)72.70 (n/a)17.22 (n/a)

eltwise_add_8_cols_2_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.09 (-45.36%)0.07 (-46.43%)0.07 (-50.88%)0.07 (-38.37%)0.01 (-57.19%)177.60 (+62.19%)166.70 (+85.22%)175.00 (+103.49%)137.80 (+83.00%)16.87 (+25.55%)
1531781 — 2026-01-23 18:49:240.16 (n/a)0.14 (n/a)0.14 (n/a)0.11 (n/a)0.02 (n/a)109.50 (n/a)90.00 (n/a)86.00 (n/a)75.30 (n/a)13.44 (n/a)

eltwise_mul_1_cols_2_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.12 (-11.59%)0.10 (-24.41%)0.10 (-28.09%)0.07 (-28.93%)0.02 (+46.16%)166.20 (+40.73%)132.46 (+35.33%)128.70 (+38.98%)102.30 (+13.16%)26.66 (+130.81%)
1531781 — 2026-01-23 18:49:240.14 (n/a)0.13 (n/a)0.13 (n/a)0.10 (n/a)0.01 (n/a)118.10 (n/a)97.88 (n/a)92.60 (n/a)90.40 (n/a)11.55 (n/a)

eltwise_mul_2_cols_2_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.11 (-21.71%)0.09 (-26.26%)0.10 (-24.86%)0.07 (-27.18%)0.02 (+3.06%)178.60 (+37.38%)140.62 (+37.73%)124.20 (+33.12%)116.30 (+27.66%)29.70 (+79.66%)
1531781 — 2026-01-23 18:49:240.13 (n/a)0.12 (n/a)0.13 (n/a)0.09 (n/a)0.02 (n/a)130.00 (n/a)102.10 (n/a)93.30 (n/a)91.10 (n/a)16.53 (n/a)

eltwise_mul_4_cols_2_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.13 (-20.62%)0.10 (-32.06%)0.09 (-37.20%)0.07 (-38.61%)0.03 (+41.70%)185.70 (+62.89%)134.76 (+56.37%)129.70 (+59.14%)93.50 (+26.01%)42.92 (+170.76%)
1531781 — 2026-01-23 18:49:240.17 (n/a)0.15 (n/a)0.15 (n/a)0.11 (n/a)0.02 (n/a)114.00 (n/a)86.18 (n/a)81.50 (n/a)74.20 (n/a)15.85 (n/a)

eltwise_mul_8_cols_2_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.07 (-42.44%)0.06 (-47.70%)0.06 (-49.00%)0.04 (-44.35%)0.01 (-27.51%)278.20 (+79.72%)218.40 (+94.51%)205.00 (+95.98%)165.90 (+73.72%)53.39 (+121.20%)
1531781 — 2026-01-23 18:49:240.13 (n/a)0.11 (n/a)0.12 (n/a)0.08 (n/a)0.02 (n/a)154.80 (n/a)112.28 (n/a)104.60 (n/a)95.50 (n/a)24.14 (n/a)

gelu_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.06 (-46.58%)0.05 (-49.93%)0.05 (-57.68%)0.04 (-50.66%)0.01 (-44.94%)206.00 (+102.76%)167.90 (+100.36%)171.20 (+136.46%)135.20 (+87.26%)32.01 (+103.22%)
1531781 — 2026-01-23 18:49:240.11 (n/a)0.10 (n/a)0.11 (n/a)0.08 (n/a)0.02 (n/a)101.60 (n/a)83.80 (n/a)72.40 (n/a)72.20 (n/a)15.75 (n/a)

gelu_1_cols_2_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.09 (-25.71%)0.06 (-45.07%)0.05 (-55.78%)0.04 (-45.42%)0.02 (-6.32%)201.50 (+83.18%)151.52 (+88.79%)156.20 (+126.05%)89.40 (+34.64%)40.83 (+120.29%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.11 (n/a)0.12 (n/a)0.07 (n/a)0.02 (n/a)110.00 (n/a)80.26 (n/a)69.10 (n/a)66.40 (n/a)18.54 (n/a)

gelu_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.07 (-38.22%)0.06 (-46.88%)0.06 (-53.25%)0.04 (-36.15%)0.01 (-49.90%)215.80 (+56.60%)155.80 (+83.25%)145.90 (+113.93%)109.70 (+61.80%)38.92 (+28.72%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.10 (n/a)0.12 (n/a)0.06 (n/a)0.03 (n/a)137.80 (n/a)85.02 (n/a)68.20 (n/a)67.80 (n/a)30.24 (n/a)

gelu_2_cols_2_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.07 (-39.64%)0.06 (-47.46%)0.05 (-53.48%)0.04 (-42.01%)0.01 (-36.67%)186.00 (+72.54%)146.68 (+90.99%)149.50 (+115.11%)111.10 (+65.57%)30.48 (+75.16%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.11 (n/a)0.12 (n/a)0.08 (n/a)0.02 (n/a)107.80 (n/a)76.80 (n/a)69.50 (n/a)67.10 (n/a)17.40 (n/a)

gelu_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.05 (-61.62%)0.05 (-56.73%)0.05 (-60.30%)0.04 (-23.21%)0.00 (-89.22%)199.30 (+30.18%)178.68 (+108.84%)174.50 (+151.80%)164.30 (+160.79%)13.16 (-65.31%)
1531781 — 2026-01-23 18:49:240.13 (n/a)0.11 (n/a)0.12 (n/a)0.05 (n/a)0.03 (n/a)153.10 (n/a)85.56 (n/a)69.30 (n/a)63.00 (n/a)37.94 (n/a)

gelu_4_cols_2_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.09 (-23.30%)0.05 (-48.26%)0.05 (-60.07%)0.04 (-50.87%)0.02 (+12.63%)224.80 (+103.44%)163.96 (+105.46%)176.40 (+150.21%)91.30 (+30.43%)49.50 (+183.14%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.11 (n/a)0.12 (n/a)0.07 (n/a)0.02 (n/a)110.50 (n/a)79.80 (n/a)70.50 (n/a)70.00 (n/a)17.48 (n/a)

gelu_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.05 (-55.37%)0.04 (-55.12%)0.05 (-60.08%)0.04 (-2.63%)0.01 (-84.70%)213.90 (+2.74%)185.08 (+87.44%)180.00 (+150.35%)157.40 (+124.22%)21.27 (-65.25%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.10 (n/a)0.11 (n/a)0.04 (n/a)0.03 (n/a)208.20 (n/a)98.74 (n/a)71.90 (n/a)70.20 (n/a)61.20 (n/a)

gelu_8_cols_2_channels_2048_tile_128

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.05 (-53.90%)0.04 (-54.26%)0.04 (-54.99%)0.04 (-42.13%)0.01 (-68.13%)230.60 (+72.86%)199.86 (+113.89%)182.90 (+122.24%)177.60 (+116.85%)26.82 (+19.39%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.09 (n/a)0.10 (n/a)0.06 (n/a)0.02 (n/a)133.40 (n/a)93.44 (n/a)82.30 (n/a)81.90 (n/a)22.47 (n/a)

gemm_1792x896x1152_64x32x48_8cols_ccolmaj

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
9369dae — 2026-02-09 20:32:364.92 (-4.41%)4.31 (-11.99%)4.37 (-12.38%)3.80 (-11.09%)0.43 (+19.84%)2476.50 (+12.48%)2199.56 (+13.97%)2151.50 (+14.14%)1910.90 (+4.61%)214.79 (+39.43%)1935.91 (-4.41%)1694.93 (-11.99%)1719.45 (-12.38%)1493.82 (-11.09%)167.75 (+19.84%)
1531781 — 2026-01-23 18:49:245.15 (n/a)4.90 (n/a)4.99 (n/a)4.27 (n/a)0.36 (n/a)2201.70 (n/a)1929.94 (n/a)1885.00 (n/a)1826.70 (n/a)154.05 (n/a)2025.19 (n/a)1925.76 (n/a)1962.51 (n/a)1680.20 (n/a)139.97 (n/a)

gemm_192x384x64_48x96x16_4cols

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
9369dae — 2026-02-09 20:32:361.16 (-50.73%)0.87 (-53.13%)0.90 (-49.55%)0.57 (-56.93%)0.23 (-42.38%)389.60 (+132.18%)270.76 (+117.93%)246.50 (+98.15%)190.30 (+103.09%)77.10 (+174.25%)49.60 (-50.73%)37.00 (-53.13%)38.28 (-49.55%)24.22 (-56.93%)9.64 (-42.38%)
1531781 — 2026-01-23 18:49:242.36 (n/a)1.85 (n/a)1.78 (n/a)1.32 (n/a)0.39 (n/a)167.80 (n/a)124.24 (n/a)124.40 (n/a)93.70 (n/a)28.11 (n/a)100.67 (n/a)78.94 (n/a)75.88 (n/a)56.25 (n/a)16.73 (n/a)

gemm_192x384x64_48x96x16_4cols_bcolmaj_ccolmaj

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
9369dae — 2026-02-09 20:32:361.40 (-36.95%)1.04 (-49.65%)1.00 (-54.38%)0.57 (-64.42%)0.31 (+17.64%)388.40 (+181.04%)234.42 (+114.67%)220.30 (+119.20%)158.30 (+58.62%)90.34 (+446.25%)59.61 (-36.95%)44.20 (-49.65%)42.84 (-54.38%)24.30 (-64.42%)13.25 (+17.64%)
1531781 — 2026-01-23 18:49:242.22 (n/a)2.06 (n/a)2.20 (n/a)1.60 (n/a)0.26 (n/a)138.20 (n/a)109.20 (n/a)100.50 (n/a)99.80 (n/a)16.54 (n/a)94.54 (n/a)87.77 (n/a)93.91 (n/a)68.30 (n/a)11.26 (n/a)

gemm_2048x2048x2048_64x64x32_8_cols_0_bcolmaj_0_ccolmaj_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)GFLOP/s (max)GFLOP/s (mean)GFLOP/s (median)GFLOP/s (min)GFLOP/s (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3a053d3 — 2025-11-26 01:12:163.72 (-0.11%)3.55 (-2.75%)3.56 (-2.75%)3.36 (-5.33%)0.12 (+76.10%)2725.23 (-0.11%)2602.52 (-2.75%)2612.11 (-2.75%)2459.89 (-5.33%)86.27 (+76.10%)6984.00 (+5.63%)6608.60 (+2.91%)6577.00 (+2.83%)6304.00 (+0.11%)221.87 (+87.15%)
9e98049 — 2025-11-25 20:50:553.72 (n/a)3.65 (n/a)3.66 (n/a)3.54 (n/a)0.07 (n/a)2728.26 (n/a)2676.06 (n/a)2686.03 (n/a)2598.29 (n/a)48.99 (n/a)6612.00 (n/a)6422.00 (n/a)6396.00 (n/a)6297.00 (n/a)118.55 (n/a)

gemm_2048x2048x2048_64x64x32_8_cols_0_bcolmaj_1_ccolmaj_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)GFLOP/s (max)GFLOP/s (mean)GFLOP/s (median)GFLOP/s (min)GFLOP/s (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3a053d3 — 2025-11-26 01:12:163.40 (-3.14%)3.38 (-0.82%)3.38 (-0.82%)3.35 (+4.93%)0.02 (-84.51%)2493.81 (-3.14%)2475.27 (-0.82%)2479.06 (-0.82%)2453.92 (+4.93%)13.33 (-84.51%)7001.00 (-4.70%)6940.80 (+0.71%)6930.00 (+0.83%)6889.00 (+3.24%)37.43 (-84.76%)
9e98049 — 2025-11-25 20:50:553.51 (n/a)3.40 (n/a)3.41 (n/a)3.19 (n/a)0.12 (n/a)2574.53 (n/a)2495.72 (n/a)2499.62 (n/a)2338.67 (n/a)86.02 (n/a)7346.00 (n/a)6892.20 (n/a)6873.00 (n/a)6673.00 (n/a)245.57 (n/a)

gemm_2048x2048x2048_64x64x32_8_cols_1_bcolmaj_0_ccolmaj_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)GFLOP/s (max)GFLOP/s (mean)GFLOP/s (median)GFLOP/s (min)GFLOP/s (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3a053d3 — 2025-11-26 01:12:163.02 (+0.52%)2.87 (-2.05%)2.86 (-2.24%)2.70 (-6.16%)0.11 (+131.66%)2216.19 (+0.52%)2104.44 (-2.05%)2094.08 (-2.24%)1980.84 (-6.16%)80.25 (+131.67%)8673.00 (+6.56%)8175.60 (+2.21%)8204.00 (+2.29%)7752.00 (-0.51%)314.03 (+145.36%)
9e98049 — 2025-11-25 20:50:553.01 (n/a)2.93 (n/a)2.92 (n/a)2.88 (n/a)0.05 (n/a)2204.81 (n/a)2148.41 (n/a)2142.13 (n/a)2110.81 (n/a)34.64 (n/a)8139.00 (n/a)7998.60 (n/a)8020.00 (n/a)7792.00 (n/a)127.99 (n/a)

gemm_2048x2048x2048_64x64x64_1cols

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
9369dae — 2026-02-09 20:32:360.52 (-0.39%)0.52 (-0.51%)0.52 (-0.44%)0.52 (-0.74%)0.00 (+235.64%)48777.30 (+0.75%)48617.40 (+0.51%)48574.70 (+0.44%)48535.10 (+0.39%)99.73 (+239.27%)353.97 (-0.39%)353.37 (-0.51%)353.68 (-0.44%)352.21 (-0.74%)0.72 (+235.64%)
1531781 — 2026-01-23 18:49:240.52 (n/a)0.52 (n/a)0.52 (n/a)0.52 (n/a)0.00 (n/a)48415.10 (n/a)48370.52 (n/a)48362.20 (n/a)48344.40 (n/a)29.40 (n/a)355.36 (n/a)355.17 (n/a)355.23 (n/a)354.85 (n/a)0.22 (n/a)

gemm_2048x2048x2048_64x64x64_2_cols_0_bcolmaj_0_ccolmaj_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:491.04 (+0.94%)1.03 (+0.23%)1.03 (+0.09%)1.02 (+0.07%)0.00 (+197.51%)24576.00 (-0.07%)24485.82 (-0.23%)24530.40 (-0.09%)24266.40 (-0.94%)111.26 (+194.35%)
0a6c11c — 2025-12-04 00:39:101.03 (n/a)1.03 (n/a)1.03 (n/a)1.02 (n/a)0.00 (n/a)24594.30 (n/a)24542.14 (n/a)24551.90 (n/a)24495.50 (n/a)37.80 (n/a)

gemm_2048x2048x2048_64x64x64_2_cols_0_bcolmaj_0_ccolmaj_0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
a4b6ffe — 2026-01-05 19:24:051.04 (+0.28%)1.03 (-0.15%)1.03 (+0.05%)1.02 (-0.52%)0.01 (+42.72%)24719.80 (+0.52%)24512.42 (+0.16%)24529.20 (-0.05%)24092.60 (-0.28%)252.92 (+43.15%)713.08 (+0.28%)700.92 (-0.15%)700.38 (+0.05%)694.98 (-0.52%)7.30 (+42.71%)
50ad37c — 2025-12-29 20:35:531.04 (n/a)1.03 (n/a)1.03 (n/a)1.02 (n/a)0.01 (n/a)24592.30 (n/a)24473.70 (n/a)24540.90 (n/a)24160.80 (n/a)176.68 (n/a)711.06 (n/a)702.00 (n/a)700.05 (n/a)698.59 (n/a)5.12 (n/a)

gemm_2048x2048x2048_64x64x64_2_cols_0_bcolmaj_1_ccolmaj_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.99 (+1.69%)0.97 (+0.29%)0.97 (-0.04%)0.97 (-0.15%)0.01 (+434.92%)26025.30 (+0.15%)25839.34 (-0.29%)25916.70 (+0.04%)25446.00 (-1.67%)202.05 (+425.44%)
0a6c11c — 2025-12-04 00:39:100.97 (n/a)0.97 (n/a)0.97 (n/a)0.97 (n/a)0.00 (n/a)25986.90 (n/a)25913.52 (n/a)25905.70 (n/a)25877.00 (n/a)38.45 (n/a)

gemm_2048x2048x2048_64x64x64_2_cols_0_bcolmaj_1_ccolmaj_0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
a4b6ffe — 2026-01-05 19:24:050.97 (-0.01%)0.97 (+0.03%)0.97 (+0.04%)0.97 (+0.00%)0.00 (-6.43%)25927.90 (-0.00%)25903.20 (-0.03%)25898.60 (-0.04%)25892.80 (+0.01%)14.05 (-6.56%)663.50 (-0.01%)663.23 (+0.03%)663.35 (+0.04%)662.60 (+0.00%)0.36 (-6.43%)
50ad37c — 2025-12-29 20:35:530.97 (n/a)0.97 (n/a)0.97 (n/a)0.97 (n/a)0.00 (n/a)25928.10 (n/a)25910.14 (n/a)25908.10 (n/a)25891.20 (n/a)15.04 (n/a)663.54 (n/a)663.06 (n/a)663.11 (n/a)662.60 (n/a)0.38 (n/a)

gemm_2048x2048x2048_64x64x64_2_cols_1_bcolmaj_0_ccolmaj_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.92 (-1.02%)0.92 (-0.66%)0.92 (-0.10%)0.91 (-0.78%)0.01 (-4.54%)27751.40 (+0.79%)27480.74 (+0.66%)27354.80 (+0.10%)27320.50 (+1.03%)182.69 (-2.84%)
0a6c11c — 2025-12-04 00:39:100.93 (n/a)0.92 (n/a)0.92 (n/a)0.91 (n/a)0.01 (n/a)27534.80 (n/a)27299.32 (n/a)27326.70 (n/a)27042.30 (n/a)188.04 (n/a)

gemm_2048x2048x2048_64x64x64_2_cols_1_bcolmaj_0_ccolmaj_0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
a4b6ffe — 2026-01-05 19:24:050.92 (+0.11%)0.91 (-0.12%)0.91 (-0.21%)0.91 (-0.24%)0.00 (+30.34%)27723.50 (+0.24%)27548.94 (+0.12%)27522.90 (+0.21%)27403.60 (-0.11%)124.26 (+30.58%)626.92 (+0.11%)623.62 (-0.12%)624.20 (-0.21%)619.69 (-0.24%)2.81 (+30.34%)
50ad37c — 2025-12-29 20:35:530.92 (n/a)0.91 (n/a)0.92 (n/a)0.91 (n/a)0.00 (n/a)27657.10 (n/a)27515.58 (n/a)27464.70 (n/a)27434.60 (n/a)95.16 (n/a)626.21 (n/a)624.37 (n/a)625.52 (n/a)621.17 (n/a)2.16 (n/a)

gemm_2048x2048x2048_64x64x64_2cols_bcolmaj

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
9369dae — 2026-02-09 20:32:360.91 (-1.48%)0.91 (-0.84%)0.91 (-1.01%)0.90 (-0.50%)0.01 (-38.81%)27966.30 (+0.50%)27767.00 (+0.85%)27799.40 (+1.02%)27511.60 (+1.50%)170.08 (-37.55%)624.46 (-1.48%)618.73 (-0.84%)618.00 (-1.01%)614.31 (-0.50%)3.80 (-38.81%)
1531781 — 2026-01-23 18:49:240.93 (n/a)0.91 (n/a)0.91 (n/a)0.90 (n/a)0.01 (n/a)27826.70 (n/a)27533.84 (n/a)27519.90 (n/a)27105.20 (n/a)272.35 (n/a)633.82 (n/a)624.00 (n/a)624.27 (n/a)617.39 (n/a)6.21 (n/a)

gemm_2048x2048x2048_64x64x64_8_cols_0_bcolmaj_0_ccolmaj_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)GFLOP/s (max)GFLOP/s (mean)GFLOP/s (median)GFLOP/s (min)GFLOP/s (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
31ab24e — 2025-11-22 02:01:563.65 (-1.06%)3.59 (+0.83%)3.63 (-0.26%)3.45 (+2.66%)0.07 (-44.71%)2678.08 (-1.06%)2632.13 (+0.83%)2658.60 (-0.26%)2528.31 (+2.66%)53.86 (-44.71%)6795.00 (-2.59%)6529.80 (-0.92%)6462.00 (+0.26%)6415.00 (+1.07%)137.14 (-45.27%)
d9a64c5 — 2025-11-21 20:49:353.69 (n/a)3.56 (n/a)3.64 (n/a)3.36 (n/a)0.13 (n/a)2706.77 (n/a)2610.50 (n/a)2665.61 (n/a)2462.71 (n/a)97.41 (n/a)6976.00 (n/a)6590.40 (n/a)6445.00 (n/a)6347.00 (n/a)250.58 (n/a)

gemm_2048x2048x2048_64x64x64_8_cols_0_bcolmaj_0_ccolmaj_0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
d48746f — 2026-01-12 22:17:103.98 (+0.20%)3.84 (-2.95%)3.76 (-5.27%)3.72 (-4.97%)0.13 (+473.97%)6760.00 (+5.23%)6559.50 (+3.13%)6697.10 (+5.56%)6321.00 (-0.20%)218.68 (+499.40%)2717.91 (+0.20%)2621.44 (-2.95%)2565.27 (-5.27%)2541.40 (-4.97%)88.37 (+473.97%)
331dcca — 2026-01-08 18:34:103.97 (n/a)3.96 (n/a)3.97 (n/a)3.92 (n/a)0.02 (n/a)6423.80 (n/a)6360.32 (n/a)6344.30 (n/a)6333.90 (n/a)36.48 (n/a)2712.39 (n/a)2701.17 (n/a)2707.91 (n/a)2674.40 (n/a)15.40 (n/a)

gemm_2048x2048x2048_64x64x64_8_cols_0_bcolmaj_1_ccolmaj_0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
d48746f — 2026-01-12 22:17:103.75 (+0.28%)3.70 (+1.25%)3.71 (-0.21%)3.65 (+4.77%)0.04 (-66.90%)6901.80 (-4.56%)6795.46 (-1.30%)6790.20 (+0.21%)6715.90 (-0.28%)67.05 (-68.40%)2558.08 (+0.28%)2528.33 (+1.25%)2530.10 (-0.21%)2489.18 (+4.77%)24.81 (-66.90%)
331dcca — 2026-01-08 18:34:103.74 (n/a)3.66 (n/a)3.71 (n/a)3.48 (n/a)0.11 (n/a)7231.20 (n/a)6885.18 (n/a)6775.70 (n/a)6734.90 (n/a)212.21 (n/a)2550.87 (n/a)2497.04 (n/a)2535.50 (n/a)2375.78 (n/a)74.96 (n/a)

gemm_2048x2048x2048_64x64x64_8_cols_1_bcolmaj_0_ccolmaj_0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
d48746f — 2026-01-12 22:17:103.62 (-0.77%)3.54 (+0.14%)3.60 (-0.04%)3.44 (+2.17%)0.09 (-25.77%)7307.40 (-2.13%)7103.46 (-0.18%)6982.50 (+0.04%)6948.20 (+0.77%)182.95 (-26.55%)2472.55 (-0.77%)2419.80 (+0.14%)2460.41 (-0.04%)2351.04 (+2.17%)61.76 (-25.77%)
331dcca — 2026-01-08 18:34:103.65 (n/a)3.54 (n/a)3.61 (n/a)3.37 (n/a)0.12 (n/a)7466.20 (n/a)7116.40 (n/a)6979.60 (n/a)6895.10 (n/a)249.09 (n/a)2491.61 (n/a)2416.45 (n/a)2461.45 (n/a)2301.00 (n/a)83.20 (n/a)

gemm_2048x2048x2048_64x64x64_8cols_bcolmaj_ccolmaj

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
9369dae — 2026-02-09 20:32:363.35 (-3.36%)3.29 (-3.75%)3.33 (-3.53%)3.18 (-3.43%)0.07 (-2.65%)7923.10 (+3.55%)7655.12 (+3.89%)7567.90 (+3.66%)7516.60 (+3.48%)169.27 (+4.15%)2285.58 (-3.36%)2245.09 (-3.75%)2270.10 (-3.53%)2168.32 (-3.43%)48.82 (-2.65%)
1531781 — 2026-01-23 18:49:243.46 (n/a)3.42 (n/a)3.45 (n/a)3.29 (n/a)0.07 (n/a)7651.20 (n/a)7368.34 (n/a)7300.80 (n/a)7263.70 (n/a)162.53 (n/a)2365.17 (n/a)2332.46 (n/a)2353.14 (n/a)2245.39 (n/a)50.15 (n/a)

gemm_384x1536x1792_32x48x64_4cols_bcolmaj

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
9369dae — 2026-02-09 20:32:364.20 (-3.54%)3.87 (-8.66%)3.82 (-11.65%)3.59 (-6.57%)0.29 (+32.50%)2242.30 (+7.04%)2090.82 (+9.72%)2107.60 (+13.19%)1920.00 (+3.67%)155.57 (+46.93%)1101.01 (-3.54%)1015.58 (-8.66%)1002.99 (-11.65%)942.74 (-6.57%)76.20 (+32.50%)
1531781 — 2026-01-23 18:49:244.35 (n/a)4.24 (n/a)4.33 (n/a)3.85 (n/a)0.22 (n/a)2094.90 (n/a)1905.64 (n/a)1862.00 (n/a)1852.00 (n/a)105.88 (n/a)1141.45 (n/a)1111.86 (n/a)1135.29 (n/a)1009.08 (n/a)57.51 (n/a)

gemm_896x1792x640_32x64x80_8cols_ccolmaj

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
9369dae — 2026-02-09 20:32:366.49 (-4.59%)5.28 (-17.93%)5.32 (-20.57%)3.65 (-31.65%)1.14 (+84.06%)1824.60 (+46.31%)1314.06 (+26.02%)1250.00 (+25.89%)1025.40 (+4.81%)320.03 (+179.28%)2004.22 (-4.59%)1631.43 (-17.93%)1644.13 (-20.57%)1126.38 (-31.65%)351.25 (+84.06%)
1531781 — 2026-01-23 18:49:246.80 (n/a)6.43 (n/a)6.70 (n/a)5.33 (n/a)0.62 (n/a)1247.10 (n/a)1042.70 (n/a)992.90 (n/a)978.30 (n/a)114.59 (n/a)2100.72 (n/a)1987.76 (n/a)2069.85 (n/a)1647.93 (n/a)190.83 (n/a)

layer_norm_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.06 (-45.58%)0.05 (-38.48%)0.06 (-31.98%)0.05 (-31.11%)0.01 (-71.87%)173.80 (+45.20%)152.78 (+56.79%)147.80 (+47.06%)131.40 (+83.78%)16.94 (-25.27%)
1531781 — 2026-01-23 18:49:240.11 (n/a)0.09 (n/a)0.08 (n/a)0.07 (n/a)0.02 (n/a)119.70 (n/a)97.44 (n/a)100.50 (n/a)71.50 (n/a)22.67 (n/a)

layer_norm_1_cols_2_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.06 (-39.98%)0.05 (-41.98%)0.05 (-43.36%)0.04 (-24.36%)0.01 (-59.48%)182.60 (+32.22%)165.40 (+67.44%)171.40 (+76.52%)127.30 (+66.62%)21.93 (-11.85%)
1531781 — 2026-01-23 18:49:240.11 (n/a)0.09 (n/a)0.08 (n/a)0.06 (n/a)0.02 (n/a)138.10 (n/a)98.78 (n/a)97.10 (n/a)76.40 (n/a)24.88 (n/a)

layer_norm_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.07 (-45.82%)0.05 (-42.65%)0.05 (-45.47%)0.04 (-26.89%)0.01 (-59.21%)191.70 (+36.83%)166.40 (+69.14%)173.10 (+83.37%)125.60 (+84.43%)26.11 (+0.00%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.09 (n/a)0.09 (n/a)0.06 (n/a)0.02 (n/a)140.10 (n/a)98.38 (n/a)94.40 (n/a)68.10 (n/a)26.11 (n/a)

layer_norm_2_cols_2_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.06 (-54.85%)0.05 (-47.40%)0.05 (-43.24%)0.04 (-35.49%)0.01 (-72.98%)202.20 (+55.06%)167.08 (+81.81%)164.70 (+76.15%)137.30 (+121.45%)23.88 (-6.66%)
1531781 — 2026-01-23 18:49:240.13 (n/a)0.09 (n/a)0.09 (n/a)0.06 (n/a)0.03 (n/a)130.40 (n/a)91.90 (n/a)93.50 (n/a)62.00 (n/a)25.59 (n/a)

layer_norm_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.06 (-44.89%)0.04 (-55.79%)0.04 (-59.62%)0.02 (-60.32%)0.01 (-35.14%)355.20 (+151.91%)217.24 (+137.32%)192.90 (+147.63%)132.60 (+81.64%)84.98 (+198.96%)
1531781 — 2026-01-23 18:49:240.11 (n/a)0.09 (n/a)0.11 (n/a)0.06 (n/a)0.02 (n/a)141.00 (n/a)91.54 (n/a)77.90 (n/a)73.00 (n/a)28.43 (n/a)

layer_norm_4_cols_2_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.05 (-57.92%)0.04 (-56.57%)0.04 (-62.43%)0.03 (-42.57%)0.01 (-76.13%)258.40 (+74.12%)224.72 (+117.88%)233.70 (+166.17%)175.20 (+137.72%)30.85 (-4.47%)
1531781 — 2026-01-23 18:49:240.11 (n/a)0.09 (n/a)0.09 (n/a)0.06 (n/a)0.02 (n/a)148.40 (n/a)103.14 (n/a)87.80 (n/a)73.70 (n/a)32.30 (n/a)

layer_norm_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.05 (-59.27%)0.05 (-53.32%)0.05 (-57.28%)0.04 (-19.10%)0.00 (-89.32%)190.30 (+23.65%)177.72 (+96.68%)181.20 (+134.11%)165.20 (+145.47%)10.87 (-69.68%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.10 (n/a)0.11 (n/a)0.05 (n/a)0.03 (n/a)153.90 (n/a)90.36 (n/a)77.40 (n/a)67.30 (n/a)35.86 (n/a)

layer_norm_8_cols_2_channels_2048_tile_128

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.05 (-44.56%)0.04 (-49.93%)0.04 (-56.71%)0.03 (-42.69%)0.01 (-45.23%)244.90 (+74.55%)212.44 (+99.47%)225.50 (+131.05%)166.70 (+80.41%)34.03 (+71.55%)
1531781 — 2026-01-23 18:49:240.09 (n/a)0.08 (n/a)0.08 (n/a)0.06 (n/a)0.01 (n/a)140.30 (n/a)106.50 (n/a)97.60 (n/a)92.40 (n/a)19.84 (n/a)

matrix_vector_mul_128x128_32_1col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:490.37 (+17.13%)0.30 (+24.87%)0.30 (+38.03%)0.23 (+12.47%)0.06 (+35.23%)
0a6c11c — 2025-12-04 00:39:100.32 (n/a)0.24 (n/a)0.22 (n/a)0.20 (n/a)0.04 (n/a)

matrix_vector_mul_128x128_32_1col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:100.26 (-1.81%)0.24 (+4.06%)0.24 (+0.52%)0.21 (+16.97%)0.02 (-48.16%)0.26 (-1.81%)0.24 (+4.06%)0.24 (+0.52%)0.20 (+16.97%)0.02 (-48.16%)
a4b6ffe — 2026-01-05 19:24:050.26 (n/a)0.23 (n/a)0.24 (n/a)0.18 (n/a)0.04 (n/a)0.26 (n/a)0.23 (n/a)0.24 (n/a)0.17 (n/a)0.04 (n/a)

matrix_vector_mul_128x128_32tsi_128tso_1col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
9369dae — 2026-02-09 20:32:360.26 (n/a)0.22 (n/a)0.23 (n/a)0.18 (n/a)0.03 (n/a)0.26 (n/a)0.22 (n/a)0.22 (n/a)0.18 (n/a)0.03 (n/a)

matrix_vector_mul_128x128_32tsi_128tso_1col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
1531781 — 2026-01-23 18:49:240.45 (+63.20%)0.35 (+43.38%)0.35 (+38.29%)0.22 (+12.34%)0.08 (+156.17%)0.44 (+63.20%)0.34 (+43.38%)0.34 (+38.29%)0.22 (+12.34%)0.08 (+156.17%)
0f0da0b — 2026-01-14 16:31:000.27 (n/a)0.24 (n/a)0.25 (n/a)0.20 (n/a)0.03 (n/a)0.27 (n/a)0.24 (n/a)0.25 (n/a)0.19 (n/a)0.03 (n/a)

matrix_vector_mul_2048x8192_1_1col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:4913.57 (+3.47%)13.15 (+1.09%)13.10 (+0.16%)12.52 (-1.58%)0.37 (+153.19%)
0a6c11c — 2025-12-04 00:39:1013.11 (n/a)13.01 (n/a)13.07 (n/a)12.72 (n/a)0.15 (n/a)

matrix_vector_mul_2048x8192_1_1col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:1013.43 (-0.07%)13.30 (+4.62%)13.27 (+1.29%)13.16 (+16.70%)0.12 (-86.69%)13.42 (-0.07%)13.29 (+4.62%)13.26 (+1.29%)13.16 (+16.70%)0.12 (-86.69%)
a4b6ffe — 2026-01-05 19:24:0513.44 (n/a)12.71 (n/a)13.10 (n/a)11.28 (n/a)0.87 (n/a)13.43 (n/a)12.70 (n/a)13.09 (n/a)11.27 (n/a)0.87 (n/a)

matrix_vector_mul_2048x8192_1_2col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:4925.06 (+2.47%)23.71 (+1.81%)24.13 (+0.50%)20.90 (+5.41%)1.48 (-15.60%)
0a6c11c — 2025-12-04 00:39:1024.46 (n/a)23.29 (n/a)24.01 (n/a)19.83 (n/a)1.75 (n/a)

matrix_vector_mul_2048x8192_1_2col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:1024.46 (-0.60%)21.07 (-8.03%)19.75 (-17.83%)18.08 (+1.18%)3.03 (+7.07%)24.44 (-0.60%)21.06 (-8.03%)19.74 (-17.83%)18.07 (+1.18%)3.03 (+7.07%)
a4b6ffe — 2026-01-05 19:24:0524.60 (n/a)22.91 (n/a)24.04 (n/a)17.87 (n/a)2.83 (n/a)24.59 (n/a)22.90 (n/a)24.02 (n/a)17.86 (n/a)2.83 (n/a)

matrix_vector_mul_2048x8192_1_4col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:4942.05 (+1.54%)40.50 (+12.60%)40.49 (+2.36%)38.98 (+97.53%)0.97 (-88.09%)
0a6c11c — 2025-12-04 00:39:1041.41 (n/a)35.97 (n/a)39.55 (n/a)19.73 (n/a)8.15 (n/a)

matrix_vector_mul_2048x8192_1_4col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:1042.58 (+1.51%)41.37 (+14.29%)41.25 (+4.89%)40.33 (+91.42%)0.92 (-89.18%)42.56 (+1.51%)41.34 (+14.29%)41.23 (+4.89%)40.30 (+91.42%)0.92 (-89.18%)
a4b6ffe — 2026-01-05 19:24:0541.95 (n/a)36.20 (n/a)39.33 (n/a)21.07 (n/a)8.54 (n/a)41.92 (n/a)36.17 (n/a)39.31 (n/a)21.05 (n/a)8.53 (n/a)

matrix_vector_mul_2048x8192_1_8col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:4946.62 (+3.11%)44.82 (+0.17%)44.78 (+0.17%)42.60 (-3.80%)1.60 (+367.72%)
0a6c11c — 2025-12-04 00:39:1045.21 (n/a)44.75 (n/a)44.70 (n/a)44.29 (n/a)0.34 (n/a)

matrix_vector_mul_2048x8192_1_8col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:1046.04 (+4.14%)45.06 (+3.47%)44.55 (+2.76%)44.30 (+2.84%)0.87 (+66.58%)46.01 (+4.14%)45.03 (+3.47%)44.52 (+2.76%)44.27 (+2.84%)0.87 (+66.58%)
a4b6ffe — 2026-01-05 19:24:0544.21 (n/a)43.55 (n/a)43.35 (n/a)43.07 (n/a)0.52 (n/a)44.18 (n/a)43.52 (n/a)43.32 (n/a)43.05 (n/a)0.52 (n/a)

matrix_vector_mul_2048x8192_1tsi_1024tso_2col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
9369dae — 2026-02-09 20:32:3624.85 (n/a)24.11 (n/a)24.33 (n/a)22.88 (n/a)0.75 (n/a)24.84 (n/a)24.10 (n/a)24.32 (n/a)22.87 (n/a)0.75 (n/a)

matrix_vector_mul_2048x8192_1tsi_1024tso_2col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
1531781 — 2026-01-23 18:49:2424.80 (+0.16%)24.59 (-0.08%)24.55 (-0.22%)24.40 (+0.04%)0.18 (+11.75%)24.79 (+0.16%)24.58 (-0.08%)24.53 (-0.22%)24.38 (+0.04%)0.18 (+11.75%)
0f0da0b — 2026-01-14 16:31:0024.76 (n/a)24.61 (n/a)24.60 (n/a)24.39 (n/a)0.16 (n/a)24.75 (n/a)24.60 (n/a)24.58 (n/a)24.37 (n/a)0.16 (n/a)

matrix_vector_mul_2048x8192_1tsi_2048tso_1col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
9369dae — 2026-02-09 20:32:3613.27 (n/a)12.46 (n/a)13.09 (n/a)10.87 (n/a)1.07 (n/a)13.26 (n/a)12.45 (n/a)13.08 (n/a)10.87 (n/a)1.07 (n/a)

matrix_vector_mul_2048x8192_1tsi_2048tso_1col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
1531781 — 2026-01-23 18:49:2413.54 (+0.77%)13.34 (+3.03%)13.45 (+0.75%)13.10 (+15.45%)0.21 (-77.02%)13.53 (+0.77%)13.33 (+3.03%)13.44 (+0.75%)13.09 (+15.45%)0.21 (-77.02%)
0f0da0b — 2026-01-14 16:31:0013.44 (n/a)12.95 (n/a)13.35 (n/a)11.35 (n/a)0.90 (n/a)13.43 (n/a)12.94 (n/a)13.34 (n/a)11.34 (n/a)0.90 (n/a)

matrix_vector_mul_2048x8192_1tsi_256tso_8col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
9369dae — 2026-02-09 20:32:3644.73 (n/a)43.29 (n/a)43.76 (n/a)40.63 (n/a)1.71 (n/a)44.71 (n/a)43.27 (n/a)43.73 (n/a)40.61 (n/a)1.71 (n/a)

matrix_vector_mul_2048x8192_1tsi_256tso_8col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
1531781 — 2026-01-23 18:49:2446.81 (+1.97%)45.91 (+2.21%)46.38 (+2.98%)44.81 (+2.36%)0.89 (-7.11%)46.79 (+1.97%)45.88 (+2.21%)46.35 (+2.98%)44.79 (+2.36%)0.89 (-7.11%)
0f0da0b — 2026-01-14 16:31:0045.91 (n/a)44.92 (n/a)45.04 (n/a)43.78 (n/a)0.96 (n/a)45.88 (n/a)44.89 (n/a)45.01 (n/a)43.75 (n/a)0.96 (n/a)

matrix_vector_mul_2048x8192_1tsi_512tso_4col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
9369dae — 2026-02-09 20:32:3641.68 (n/a)39.41 (n/a)39.60 (n/a)37.18 (n/a)2.00 (n/a)41.65 (n/a)39.38 (n/a)39.58 (n/a)37.15 (n/a)2.00 (n/a)

matrix_vector_mul_2048x8192_1tsi_512tso_4col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
1531781 — 2026-01-23 18:49:2441.49 (-0.92%)40.81 (+0.09%)41.13 (+0.38%)39.95 (+0.86%)0.65 (-24.40%)41.46 (-0.92%)40.79 (+0.09%)41.10 (+0.38%)39.93 (+0.86%)0.65 (-24.40%)
0f0da0b — 2026-01-14 16:31:0041.88 (n/a)40.78 (n/a)40.97 (n/a)39.61 (n/a)0.86 (n/a)41.85 (n/a)40.75 (n/a)40.95 (n/a)39.59 (n/a)0.86 (n/a)

matrix_vector_mul_8192x2048_4_1col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:4913.29 (+1.63%)13.19 (+6.08%)13.19 (+1.17%)13.09 (+26.86%)0.08 (-92.94%)
0a6c11c — 2025-12-04 00:39:1013.08 (n/a)12.44 (n/a)13.03 (n/a)10.32 (n/a)1.07 (n/a)

matrix_vector_mul_8192x2048_4_1col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:1013.68 (+3.42%)13.30 (+8.54%)13.38 (+7.25%)12.72 (+14.96%)0.35 (-66.09%)13.67 (+3.42%)13.29 (+8.54%)13.38 (+7.25%)12.71 (+14.96%)0.35 (-66.09%)
a4b6ffe — 2026-01-05 19:24:0513.23 (n/a)12.25 (n/a)12.48 (n/a)11.07 (n/a)1.04 (n/a)13.22 (n/a)12.24 (n/a)12.47 (n/a)11.06 (n/a)1.04 (n/a)

matrix_vector_mul_8192x2048_4_2col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:4925.58 (+4.24%)24.23 (+6.59%)24.03 (+0.29%)23.58 (+37.29%)0.70 (-74.97%)
0a6c11c — 2025-12-04 00:39:1024.54 (n/a)22.73 (n/a)23.96 (n/a)17.18 (n/a)2.79 (n/a)

matrix_vector_mul_8192x2048_4_2col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:1025.08 (+1.32%)24.68 (+13.42%)24.61 (+3.26%)24.47 (+48.32%)0.24 (-93.56%)25.07 (+1.32%)24.67 (+13.42%)24.59 (+3.26%)24.46 (+48.32%)0.24 (-93.56%)
a4b6ffe — 2026-01-05 19:24:0524.76 (n/a)21.76 (n/a)23.83 (n/a)16.50 (n/a)3.74 (n/a)24.74 (n/a)21.75 (n/a)23.82 (n/a)16.49 (n/a)3.73 (n/a)

matrix_vector_mul_8192x2048_4_4col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:4940.94 (-2.19%)39.74 (-1.10%)39.76 (-0.60%)38.17 (-2.14%)1.04 (+4.39%)
0a6c11c — 2025-12-04 00:39:1041.86 (n/a)40.18 (n/a)40.00 (n/a)39.00 (n/a)1.00 (n/a)

matrix_vector_mul_8192x2048_4_4col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:1040.74 (-1.09%)37.16 (-7.15%)40.25 (+1.47%)24.51 (-37.44%)7.07 (+736.13%)40.72 (-1.09%)37.13 (-7.15%)40.23 (+1.47%)24.50 (-37.44%)7.07 (+736.13%)
a4b6ffe — 2026-01-05 19:24:0541.19 (n/a)40.02 (n/a)39.67 (n/a)39.18 (n/a)0.85 (n/a)41.17 (n/a)39.99 (n/a)39.64 (n/a)39.16 (n/a)0.85 (n/a)

matrix_vector_mul_8192x2048_4_8col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:4946.71 (-1.04%)44.91 (-2.34%)44.71 (-2.98%)43.80 (-1.44%)0.96 (+6.93%)
0a6c11c — 2025-12-04 00:39:1047.20 (n/a)45.98 (n/a)46.08 (n/a)44.44 (n/a)0.90 (n/a)

matrix_vector_mul_8192x2048_4_8col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:1046.28 (+5.68%)44.89 (+14.59%)44.74 (+4.26%)44.20 (+62.80%)0.84 (-87.96%)46.25 (+5.68%)44.86 (+14.59%)44.71 (+4.26%)44.17 (+62.80%)0.84 (-87.96%)
a4b6ffe — 2026-01-05 19:24:0543.79 (n/a)39.17 (n/a)42.91 (n/a)27.15 (n/a)6.98 (n/a)43.77 (n/a)39.15 (n/a)42.88 (n/a)27.13 (n/a)6.98 (n/a)

matrix_vector_mul_8192x2048_4tsi_1024tso_1col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
9369dae — 2026-02-09 20:32:3613.52 (n/a)13.26 (n/a)13.24 (n/a)13.08 (n/a)0.17 (n/a)13.51 (n/a)13.25 (n/a)13.23 (n/a)13.07 (n/a)0.17 (n/a)

matrix_vector_mul_8192x2048_4tsi_1024tso_1col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
1531781 — 2026-01-23 18:49:2413.46 (+0.53%)13.33 (+2.85%)13.32 (+1.17%)13.20 (+10.91%)0.11 (-81.73%)13.45 (+0.53%)13.32 (+2.85%)13.31 (+1.17%)13.19 (+10.91%)0.11 (-81.73%)
0f0da0b — 2026-01-14 16:31:0013.39 (n/a)12.96 (n/a)13.17 (n/a)11.90 (n/a)0.62 (n/a)13.38 (n/a)12.95 (n/a)13.16 (n/a)11.89 (n/a)0.62 (n/a)

matrix_vector_mul_8192x2048_4tsi_1024tso_2col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
9369dae — 2026-02-09 20:32:3625.17 (n/a)24.32 (n/a)24.41 (n/a)23.40 (n/a)0.64 (n/a)25.15 (n/a)24.31 (n/a)24.40 (n/a)23.39 (n/a)0.64 (n/a)

matrix_vector_mul_8192x2048_4tsi_1024tso_2col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
1531781 — 2026-01-23 18:49:2425.27 (+0.55%)24.75 (-0.75%)24.61 (-1.28%)24.55 (-0.37%)0.30 (+52.22%)25.25 (+0.55%)24.74 (-0.75%)24.59 (-1.28%)24.53 (-0.37%)0.30 (+52.22%)
0f0da0b — 2026-01-14 16:31:0025.13 (n/a)24.94 (n/a)24.93 (n/a)24.64 (n/a)0.20 (n/a)25.11 (n/a)24.92 (n/a)24.91 (n/a)24.62 (n/a)0.20 (n/a)

matrix_vector_mul_8192x2048_4tsi_1024tso_4col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
9369dae — 2026-02-09 20:32:3641.85 (n/a)39.89 (n/a)40.04 (n/a)38.63 (n/a)1.32 (n/a)41.82 (n/a)39.87 (n/a)40.02 (n/a)38.61 (n/a)1.32 (n/a)

matrix_vector_mul_8192x2048_4tsi_1024tso_4col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
1531781 — 2026-01-23 18:49:2440.86 (-4.37%)40.01 (-4.03%)40.43 (-3.32%)38.86 (-2.46%)0.89 (-24.84%)40.84 (-4.37%)39.98 (-4.03%)40.41 (-3.32%)38.84 (-2.46%)0.89 (-24.84%)
0f0da0b — 2026-01-14 16:31:0042.73 (n/a)41.69 (n/a)41.82 (n/a)39.85 (n/a)1.18 (n/a)42.70 (n/a)41.66 (n/a)41.80 (n/a)39.82 (n/a)1.18 (n/a)

matrix_vector_mul_8192x2048_4tsi_1024tso_8col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
9369dae — 2026-02-09 20:32:3644.51 (n/a)42.95 (n/a)42.24 (n/a)41.36 (n/a)1.44 (n/a)44.48 (n/a)42.93 (n/a)42.21 (n/a)41.33 (n/a)1.44 (n/a)

matrix_vector_mul_8192x2048_4tsi_1024tso_8col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
1531781 — 2026-01-23 18:49:2446.08 (-1.82%)45.11 (-1.49%)45.49 (-0.25%)43.89 (-1.05%)0.97 (-6.46%)46.05 (-1.82%)45.08 (-1.49%)45.46 (-0.25%)43.87 (-1.05%)0.97 (-6.46%)
0f0da0b — 2026-01-14 16:31:0046.94 (n/a)45.79 (n/a)45.60 (n/a)44.36 (n/a)1.03 (n/a)46.91 (n/a)45.76 (n/a)45.58 (n/a)44.33 (n/a)1.03 (n/a)

mem_copy_16_cores_2_chans_2048_tile_128_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.06 (-13.16%)0.05 (-19.49%)0.05 (-17.62%)0.04 (-11.92%)0.01 (-5.26%)196.70 (+13.50%)171.18 (+24.10%)162.10 (+21.33%)142.10 (+15.15%)23.10 (+25.50%)
130b6ea — 2025-12-05 21:15:490.07 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)173.30 (n/a)137.94 (n/a)133.60 (n/a)123.40 (n/a)18.40 (n/a)

mem_copy_16_cores_2_chans_2048_tile_128_False0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.10 (+63.55%)0.08 (+85.69%)0.09 (+119.34%)0.04 (+23.15%)0.02 (+126.16%)195.30 (-18.79%)111.62 (-42.82%)86.60 (-54.42%)84.60 (-38.87%)47.65 (+9.68%)
0f0da0b — 2026-01-14 16:31:000.06 (n/a)0.04 (n/a)0.04 (n/a)0.03 (n/a)0.01 (n/a)240.50 (n/a)195.20 (n/a)190.00 (n/a)138.40 (n/a)43.45 (n/a)

mem_copy_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
003b49d — 2025-11-14 19:30:540.05 (-16.99%)0.05 (-12.47%)0.05 (-7.32%)0.04 (-11.17%)0.00 (-29.61%)197.00 (+12.57%)171.00 (+13.70%)164.00 (+7.89%)153.00 (+20.47%)17.49 (-4.32%)
0999a9a — 2025-11-13 19:24:080.06 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)175.00 (n/a)150.40 (n/a)152.00 (n/a)127.00 (n/a)18.28 (n/a)

mem_copy_1_cols_2_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
003b49d — 2025-11-14 19:30:540.06 (+5.56%)0.05 (-1.59%)0.05 (-1.39%)0.03 (-14.07%)0.01 (+25.03%)263.00 (+16.37%)166.00 (+4.53%)144.00 (+1.41%)126.00 (-5.26%)49.38 (+43.17%)
0999a9a — 2025-11-13 19:24:080.06 (n/a)0.05 (n/a)0.05 (n/a)0.03 (n/a)0.01 (n/a)226.00 (n/a)158.80 (n/a)142.00 (n/a)133.00 (n/a)34.49 (n/a)

mem_copy_1_cores_1_chans_2048_tile_2048_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.07 (-21.63%)0.06 (-15.57%)0.06 (-9.81%)0.05 (-12.84%)0.01 (-28.56%)172.40 (+14.78%)142.80 (+17.36%)139.20 (+10.83%)124.90 (+27.58%)19.54 (+5.31%)
130b6ea — 2025-12-05 21:15:490.08 (n/a)0.07 (n/a)0.07 (n/a)0.05 (n/a)0.01 (n/a)150.20 (n/a)121.68 (n/a)125.60 (n/a)97.90 (n/a)18.56 (n/a)

mem_copy_1_cores_1_chans_2048_tile_2048_False0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.10 (+31.19%)0.09 (+26.03%)0.10 (+33.11%)0.07 (+25.01%)0.02 (+78.79%)118.10 (-20.04%)96.52 (-19.61%)84.40 (-24.91%)82.40 (-23.77%)18.29 (+9.09%)
0f0da0b — 2026-01-14 16:31:000.08 (n/a)0.07 (n/a)0.07 (n/a)0.06 (n/a)0.01 (n/a)147.70 (n/a)120.06 (n/a)112.40 (n/a)108.10 (n/a)16.76 (n/a)

mem_copy_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
003b49d — 2025-11-14 19:30:540.06 (-15.32%)0.05 (-17.09%)0.05 (-23.81%)0.04 (-15.54%)0.01 (-18.18%)193.00 (+18.40%)160.80 (+20.36%)168.00 (+31.25%)124.00 (+18.10%)22.92 (+11.39%)
0999a9a — 2025-11-13 19:24:080.07 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)163.00 (n/a)133.60 (n/a)128.00 (n/a)105.00 (n/a)20.58 (n/a)

mem_copy_2_cols_2_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
003b49d — 2025-11-14 19:30:540.06 (-23.02%)0.05 (-6.98%)0.05 (-2.13%)0.04 (-15.31%)0.01 (-33.29%)196.00 (+18.07%)147.00 (+6.52%)141.00 (+2.17%)126.00 (+29.90%)25.27 (+8.13%)
0999a9a — 2025-11-13 19:24:080.08 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)166.00 (n/a)138.00 (n/a)138.00 (n/a)97.00 (n/a)23.37 (n/a)

mem_copy_2_cores_1_chans_2048_tile_1024_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.07 (-28.54%)0.06 (-24.48%)0.07 (-12.25%)0.05 (-38.06%)0.01 (+26.77%)177.40 (+61.42%)133.28 (+35.58%)114.90 (+13.99%)111.10 (+39.92%)29.89 (+183.59%)
130b6ea — 2025-12-05 21:15:490.10 (n/a)0.08 (n/a)0.08 (n/a)0.07 (n/a)0.01 (n/a)109.90 (n/a)98.30 (n/a)100.80 (n/a)79.40 (n/a)10.54 (n/a)

mem_copy_2_cores_1_chans_2048_tile_1024_False0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.12 (+103.55%)0.09 (+81.99%)0.09 (+69.86%)0.07 (+83.95%)0.02 (+187.42%)117.10 (-45.66%)92.20 (-43.53%)86.90 (-41.12%)70.40 (-50.84%)23.32 (-23.43%)
0f0da0b — 2026-01-14 16:31:000.06 (n/a)0.05 (n/a)0.06 (n/a)0.04 (n/a)0.01 (n/a)215.50 (n/a)163.26 (n/a)147.60 (n/a)143.20 (n/a)30.45 (n/a)

mem_copy_2_cores_2_chans_2048_tile_1024_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.08 (-14.81%)0.07 (-11.21%)0.06 (-11.59%)0.05 (-1.47%)0.01 (-14.59%)152.30 (+1.47%)127.84 (+11.55%)126.40 (+13.16%)98.50 (+17.40%)21.87 (+3.00%)
130b6ea — 2025-12-05 21:15:490.10 (n/a)0.07 (n/a)0.07 (n/a)0.05 (n/a)0.01 (n/a)150.10 (n/a)114.60 (n/a)111.70 (n/a)83.90 (n/a)21.23 (n/a)

mem_copy_2_cores_2_chans_2048_tile_1024_False0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.10 (+35.25%)0.08 (+37.83%)0.09 (+48.91%)0.06 (+9.50%)0.02 (+125.91%)140.20 (-8.66%)101.66 (-25.39%)92.10 (-32.82%)82.90 (-26.11%)24.23 (+51.17%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)153.50 (n/a)136.26 (n/a)137.10 (n/a)112.20 (n/a)16.03 (n/a)

mem_copy_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
003b49d — 2025-11-14 19:30:540.07 (+11.97%)0.05 (+3.09%)0.05 (+2.10%)0.03 (-11.35%)0.01 (+41.65%)229.00 (+12.81%)156.00 (-0.51%)143.00 (-2.05%)117.00 (-10.69%)38.21 (+48.71%)
0999a9a — 2025-11-13 19:24:080.06 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)203.00 (n/a)156.80 (n/a)146.00 (n/a)131.00 (n/a)25.69 (n/a)

mem_copy_4_cols_2_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
003b49d — 2025-11-14 19:30:540.06 (+9.02%)0.05 (+5.59%)0.05 (+0.00%)0.04 (+3.63%)0.01 (+25.48%)193.00 (-3.50%)147.40 (-4.66%)147.00 (+0.00%)122.00 (-8.27%)25.24 (+7.48%)
0999a9a — 2025-11-13 19:24:080.06 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)200.00 (n/a)154.60 (n/a)147.00 (n/a)133.00 (n/a)23.48 (n/a)

mem_copy_4_cores_1_chans_2048_tile_512_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.07 (-18.30%)0.06 (-21.43%)0.05 (-21.71%)0.05 (-21.48%)0.01 (-9.76%)165.20 (+27.37%)144.10 (+27.18%)149.30 (+27.72%)110.60 (+22.35%)20.25 (+34.69%)
130b6ea — 2025-12-05 21:15:490.09 (n/a)0.07 (n/a)0.07 (n/a)0.06 (n/a)0.01 (n/a)129.70 (n/a)113.30 (n/a)116.90 (n/a)90.40 (n/a)15.04 (n/a)

mem_copy_4_cores_1_chans_2048_tile_512_False0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.12 (+51.68%)0.10 (+55.58%)0.11 (+85.86%)0.06 (+8.92%)0.03 (+163.31%)128.30 (-8.23%)84.56 (-32.56%)71.60 (-46.17%)65.70 (-34.04%)26.42 (+57.86%)
0f0da0b — 2026-01-14 16:31:000.08 (n/a)0.07 (n/a)0.06 (n/a)0.06 (n/a)0.01 (n/a)139.80 (n/a)125.38 (n/a)133.00 (n/a)99.60 (n/a)16.73 (n/a)

mem_copy_4_cores_2_chans_2048_tile_512_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.07 (-21.28%)0.06 (-16.50%)0.06 (-10.41%)0.05 (-22.94%)0.01 (-21.43%)174.50 (+29.74%)139.24 (+19.38%)130.40 (+11.64%)124.50 (+27.04%)20.53 (+30.67%)
130b6ea — 2025-12-05 21:15:490.08 (n/a)0.07 (n/a)0.07 (n/a)0.06 (n/a)0.01 (n/a)134.50 (n/a)116.64 (n/a)116.80 (n/a)98.00 (n/a)15.71 (n/a)

mem_copy_4_cores_2_chans_2048_tile_512_False0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.10 (+21.52%)0.09 (+51.62%)0.10 (+79.69%)0.06 (+23.77%)0.02 (+25.48%)133.20 (-19.17%)92.98 (-33.88%)83.40 (-44.33%)82.10 (-17.74%)22.49 (-15.24%)
0f0da0b — 2026-01-14 16:31:000.08 (n/a)0.06 (n/a)0.05 (n/a)0.05 (n/a)0.01 (n/a)164.80 (n/a)140.62 (n/a)149.80 (n/a)99.80 (n/a)26.53 (n/a)

mem_copy_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
003b49d — 2025-11-14 19:30:540.05 (-13.48%)0.05 (-17.79%)0.05 (-15.33%)0.03 (-25.09%)0.01 (+6.67%)267.00 (+33.50%)176.80 (+24.51%)150.00 (+18.11%)141.00 (+15.57%)47.41 (+61.41%)
0999a9a — 2025-11-13 19:24:080.06 (n/a)0.06 (n/a)0.06 (n/a)0.04 (n/a)0.01 (n/a)200.00 (n/a)142.00 (n/a)127.00 (n/a)122.00 (n/a)29.37 (n/a)

mem_copy_8_cols_2_channels_2048_tile_128

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
003b49d — 2025-11-14 19:30:540.05 (+13.38%)0.05 (+23.85%)0.05 (+37.91%)0.05 (+30.72%)0.00 (-56.85%)166.00 (-23.50%)153.40 (-20.60%)153.00 (-27.49%)142.00 (-11.80%)7.63 (-70.60%)
0999a9a — 2025-11-13 19:24:080.05 (n/a)0.04 (n/a)0.04 (n/a)0.04 (n/a)0.01 (n/a)217.00 (n/a)193.20 (n/a)211.00 (n/a)161.00 (n/a)25.96 (n/a)

mem_copy_8_cores_1_chans_2048_tile_256_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.06 (-25.56%)0.05 (-33.73%)0.05 (-35.79%)0.04 (-44.10%)0.01 (+54.66%)209.80 (+78.86%)166.22 (+54.19%)169.00 (+55.76%)126.40 (+34.33%)30.62 (+269.13%)
130b6ea — 2025-12-05 21:15:490.09 (n/a)0.08 (n/a)0.08 (n/a)0.07 (n/a)0.01 (n/a)117.30 (n/a)107.80 (n/a)108.50 (n/a)94.10 (n/a)8.29 (n/a)

mem_copy_8_cores_1_chans_2048_tile_256_False0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.09 (+46.69%)0.08 (+57.18%)0.09 (+64.02%)0.06 (+94.45%)0.01 (+6.58%)131.30 (-48.59%)99.78 (-38.77%)88.70 (-39.00%)86.90 (-31.84%)18.90 (-64.18%)
0f0da0b — 2026-01-14 16:31:000.06 (n/a)0.05 (n/a)0.06 (n/a)0.03 (n/a)0.01 (n/a)255.40 (n/a)162.96 (n/a)145.40 (n/a)127.50 (n/a)52.78 (n/a)

mem_copy_8_cores_2_chans_2048_tile_256_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.06 (-30.55%)0.05 (-25.61%)0.05 (-25.56%)0.05 (-21.98%)0.01 (-40.98%)175.00 (+28.21%)160.92 (+33.15%)168.10 (+34.37%)131.80 (+43.89%)17.33 (+10.25%)
130b6ea — 2025-12-05 21:15:490.09 (n/a)0.07 (n/a)0.07 (n/a)0.06 (n/a)0.01 (n/a)136.50 (n/a)120.86 (n/a)125.10 (n/a)91.60 (n/a)15.72 (n/a)

mem_copy_8_cores_2_chans_2048_tile_256_False0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.10 (+63.97%)0.09 (+58.77%)0.09 (+53.52%)0.06 (+50.07%)0.02 (+70.34%)144.40 (-33.36%)97.70 (-36.49%)88.10 (-34.89%)80.00 (-39.02%)26.35 (-27.55%)
0f0da0b — 2026-01-14 16:31:000.06 (n/a)0.06 (n/a)0.06 (n/a)0.04 (n/a)0.01 (n/a)216.70 (n/a)153.84 (n/a)135.30 (n/a)131.20 (n/a)36.36 (n/a)

mha

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.18 (-0.57%)0.18 (-0.76%)0.18 (-0.69%)0.18 (-0.99%)0.00 (+289.77%)47728.80 (+1.00%)47575.70 (+0.77%)47540.40 (+0.70%)47436.20 (+0.57%)113.73 (+296.19%)
130b6ea — 2025-12-05 21:15:490.18 (n/a)0.18 (n/a)0.18 (n/a)0.18 (n/a)0.00 (n/a)47253.90 (n/a)47213.70 (n/a)47210.90 (n/a)47167.70 (n/a)28.71 (n/a)

mha0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.18 (-0.08%)0.18 (+0.22%)0.18 (+0.32%)0.18 (+0.35%)0.00 (-42.58%)47367.50 (-0.34%)47232.68 (-0.22%)47232.00 (-0.32%)47073.00 (+0.08%)105.45 (-42.71%)
0f0da0b — 2026-01-14 16:31:000.18 (n/a)0.18 (n/a)0.18 (n/a)0.18 (n/a)0.00 (n/a)47531.20 (n/a)47335.18 (n/a)47384.30 (n/a)47037.20 (n/a)184.08 (n/a)

relu_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.07 (-38.74%)0.05 (-41.75%)0.05 (-51.12%)0.04 (-23.64%)0.01 (-61.28%)200.30 (+31.00%)164.32 (+61.70%)168.10 (+104.50%)119.20 (+63.29%)29.25 (-18.18%)
1531781 — 2026-01-23 18:49:240.11 (n/a)0.09 (n/a)0.10 (n/a)0.05 (n/a)0.03 (n/a)152.90 (n/a)101.62 (n/a)82.20 (n/a)73.00 (n/a)35.75 (n/a)

relu_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.07 (-32.49%)0.05 (-35.36%)0.05 (-43.07%)0.05 (-11.81%)0.01 (-56.72%)170.70 (+13.35%)151.64 (+49.75%)160.50 (+75.79%)123.00 (+48.19%)19.63 (-29.95%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.08 (n/a)0.09 (n/a)0.05 (n/a)0.02 (n/a)150.60 (n/a)101.26 (n/a)91.30 (n/a)83.00 (n/a)28.02 (n/a)

relu_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.07 (-27.40%)0.06 (-35.75%)0.05 (-44.68%)0.05 (-25.32%)0.01 (-20.35%)181.90 (+33.95%)150.52 (+56.47%)155.90 (+80.65%)114.40 (+37.67%)32.65 (+45.39%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.09 (n/a)0.09 (n/a)0.06 (n/a)0.02 (n/a)135.80 (n/a)96.20 (n/a)86.30 (n/a)83.10 (n/a)22.46 (n/a)

relu_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.05 (-51.61%)0.05 (-49.60%)0.05 (-52.23%)0.04 (-22.81%)0.00 (-80.07%)200.00 (+29.53%)173.06 (+85.45%)164.00 (+109.18%)156.60 (+106.60%)17.96 (-47.55%)
1531781 — 2026-01-23 18:49:240.11 (n/a)0.09 (n/a)0.10 (n/a)0.05 (n/a)0.02 (n/a)154.40 (n/a)93.32 (n/a)78.40 (n/a)75.80 (n/a)34.24 (n/a)

rms_norm_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.06 (-30.34%)0.05 (-33.99%)0.06 (-37.80%)0.04 (-27.80%)0.01 (-33.29%)182.20 (+38.56%)152.10 (+51.04%)148.50 (+60.71%)130.40 (+43.45%)22.40 (+29.14%)
1531781 — 2026-01-23 18:49:240.09 (n/a)0.08 (n/a)0.09 (n/a)0.06 (n/a)0.01 (n/a)131.50 (n/a)100.70 (n/a)92.40 (n/a)90.90 (n/a)17.35 (n/a)

rms_norm_1_cols_2_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.06 (-46.97%)0.05 (-46.40%)0.05 (-52.33%)0.05 (-10.57%)0.00 (-81.94%)162.40 (+11.85%)149.98 (+73.31%)151.20 (+109.71%)133.00 (+88.65%)12.38 (-62.25%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.10 (n/a)0.11 (n/a)0.06 (n/a)0.03 (n/a)145.20 (n/a)86.54 (n/a)72.10 (n/a)70.50 (n/a)32.80 (n/a)

rms_norm_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.06 (-32.50%)0.05 (-37.14%)0.05 (-39.79%)0.05 (-25.43%)0.01 (-55.32%)173.80 (+34.10%)154.28 (+57.08%)150.90 (+66.01%)133.00 (+48.11%)15.50 (-11.70%)
1531781 — 2026-01-23 18:49:240.09 (n/a)0.09 (n/a)0.09 (n/a)0.06 (n/a)0.01 (n/a)129.60 (n/a)98.22 (n/a)90.90 (n/a)89.80 (n/a)17.55 (n/a)

rms_norm_2_cols_2_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.05 (-60.53%)0.04 (-62.30%)0.05 (-57.29%)0.03 (-71.47%)0.01 (-31.81%)327.60 (+250.37%)207.82 (+179.63%)171.20 (+134.20%)158.10 (+153.37%)69.99 (+503.19%)
1531781 — 2026-01-23 18:49:240.13 (n/a)0.11 (n/a)0.11 (n/a)0.09 (n/a)0.02 (n/a)93.50 (n/a)74.32 (n/a)73.10 (n/a)62.40 (n/a)11.60 (n/a)

rms_norm_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.06 (-36.15%)0.05 (-46.86%)0.04 (-53.34%)0.04 (-37.68%)0.01 (-34.77%)209.80 (+60.52%)183.44 (+88.18%)193.40 (+114.41%)134.70 (+56.63%)28.63 (+53.38%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.09 (n/a)0.09 (n/a)0.06 (n/a)0.01 (n/a)130.70 (n/a)97.48 (n/a)90.20 (n/a)86.00 (n/a)18.66 (n/a)

rms_norm_4_cols_2_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.06 (-42.18%)0.05 (-50.65%)0.06 (-50.46%)0.03 (-47.87%)0.02 (-26.64%)240.80 (+91.87%)176.18 (+109.99%)148.30 (+101.77%)126.10 (+72.98%)57.68 (+147.96%)
1531781 — 2026-01-23 18:49:240.11 (n/a)0.10 (n/a)0.11 (n/a)0.07 (n/a)0.02 (n/a)125.50 (n/a)83.90 (n/a)73.50 (n/a)72.90 (n/a)23.26 (n/a)

rms_norm_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.07 (-17.48%)0.05 (-37.42%)0.05 (-42.78%)0.03 (-40.40%)0.01 (-3.35%)258.90 (+67.79%)175.68 (+64.71%)165.70 (+74.79%)114.00 (+21.28%)52.55 (+97.28%)
1531781 — 2026-01-23 18:49:240.09 (n/a)0.08 (n/a)0.09 (n/a)0.05 (n/a)0.01 (n/a)154.30 (n/a)106.66 (n/a)94.80 (n/a)94.00 (n/a)26.64 (n/a)

rms_norm_8_cols_2_channels_2048_tile_128

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.05 (-46.72%)0.04 (-48.25%)0.04 (-58.25%)0.03 (-30.64%)0.01 (-62.29%)241.80 (+44.19%)203.90 (+86.45%)219.30 (+139.41%)166.90 (+87.74%)33.02 (-1.50%)
1531781 — 2026-01-23 18:49:240.09 (n/a)0.08 (n/a)0.09 (n/a)0.05 (n/a)0.02 (n/a)167.70 (n/a)109.36 (n/a)91.60 (n/a)88.90 (n/a)33.52 (n/a)

rope_1_cols_2_channels_4096_tile_4096_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1b2d230 — 2026-01-13 22:38:030.19 (-21.66%)0.17 (-8.53%)0.18 (+2.09%)0.14 (-6.71%)0.02 (-38.07%)174.10 (+7.20%)150.06 (+8.11%)139.60 (-2.04%)129.80 (+27.63%)20.36 (-10.65%)
d48746f — 2026-01-12 22:17:100.24 (n/a)0.18 (n/a)0.17 (n/a)0.15 (n/a)0.04 (n/a)162.40 (n/a)138.80 (n/a)142.50 (n/a)101.70 (n/a)22.79 (n/a)

rope_1c_32rows_512cols_32arows_0m

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.79 (-27.99%)0.65 (-34.61%)0.63 (-40.41%)0.43 (-41.43%)0.15 (-3.48%)228.40 (+70.70%)159.24 (+56.82%)156.60 (+67.85%)123.70 (+38.83%)41.96 (+125.61%)
1531781 — 2026-01-23 18:49:241.10 (n/a)0.99 (n/a)1.05 (n/a)0.73 (n/a)0.15 (n/a)133.80 (n/a)101.54 (n/a)93.30 (n/a)89.10 (n/a)18.60 (n/a)

rope_1c_32rows_512cols_8arows_0m

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.60 (+7.90%)0.49 (-7.28%)0.50 (-5.50%)0.35 (-27.32%)0.09 (+196.84%)211.50 (+37.61%)156.32 (+10.99%)148.30 (+5.85%)122.80 (-7.32%)33.04 (+293.33%)
1531781 — 2026-01-23 18:49:240.56 (n/a)0.52 (n/a)0.53 (n/a)0.48 (n/a)0.03 (n/a)153.70 (n/a)140.84 (n/a)140.10 (n/a)132.50 (n/a)8.40 (n/a)

rope_2_cols_2_channels_4096_tile_2048_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1b2d230 — 2026-01-13 22:38:030.21 (-0.56%)0.17 (-10.62%)0.16 (-13.58%)0.13 (-13.33%)0.03 (+17.05%)183.70 (+15.39%)151.76 (+12.77%)155.20 (+15.73%)117.30 (+0.60%)24.24 (+35.73%)
d48746f — 2026-01-12 22:17:100.21 (n/a)0.19 (n/a)0.18 (n/a)0.15 (n/a)0.02 (n/a)159.20 (n/a)134.58 (n/a)134.10 (n/a)116.60 (n/a)17.86 (n/a)

rope_2c_32rows_512cols_32arows_0m

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.89 (-28.54%)0.62 (-35.71%)0.56 (-37.77%)0.51 (-17.81%)0.15 (-38.54%)191.00 (+21.66%)164.54 (+52.21%)174.00 (+60.66%)110.10 (+39.90%)31.51 (+1.33%)
1531781 — 2026-01-23 18:49:241.25 (n/a)0.97 (n/a)0.91 (n/a)0.63 (n/a)0.25 (n/a)157.00 (n/a)108.10 (n/a)108.30 (n/a)78.70 (n/a)31.09 (n/a)

rope_2c_32rows_512cols_8arows_0m

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.66 (+4.88%)0.45 (-20.36%)0.43 (-24.38%)0.32 (-34.14%)0.13 (+153.85%)228.60 (+51.79%)173.04 (+32.52%)172.40 (+32.21%)111.30 (-4.71%)45.08 (+258.75%)
1531781 — 2026-01-23 18:49:240.63 (n/a)0.57 (n/a)0.57 (n/a)0.49 (n/a)0.05 (n/a)150.60 (n/a)130.58 (n/a)130.40 (n/a)116.80 (n/a)12.57 (n/a)

rope_4_cols_2_channels_4096_tile_1024_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1b2d230 — 2026-01-13 22:38:030.22 (+18.92%)0.17 (+6.24%)0.17 (+5.57%)0.14 (-0.50%)0.03 (+90.77%)171.10 (+0.47%)147.40 (-4.39%)148.10 (-5.31%)110.10 (-15.89%)23.59 (+60.67%)
d48746f — 2026-01-12 22:17:100.19 (n/a)0.16 (n/a)0.16 (n/a)0.14 (n/a)0.02 (n/a)170.30 (n/a)154.16 (n/a)156.40 (n/a)130.90 (n/a)14.69 (n/a)

rope_8_cols_2_channels_4096_tile_512_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1b2d230 — 2026-01-13 22:38:030.17 (-1.30%)0.15 (+6.63%)0.15 (-8.78%)0.14 (+39.68%)0.01 (-67.35%)174.90 (-28.41%)163.42 (-10.36%)167.40 (+9.63%)146.50 (+1.31%)11.19 (-76.17%)
d48746f — 2026-01-12 22:17:100.17 (n/a)0.14 (n/a)0.16 (n/a)0.10 (n/a)0.03 (n/a)244.30 (n/a)182.30 (n/a)152.70 (n/a)144.60 (n/a)46.98 (n/a)

rope_8c_32rows_512cols_32arows_0m

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.79 (-40.55%)0.57 (-36.75%)0.54 (-36.59%)0.42 (-30.09%)0.14 (-47.60%)233.90 (+43.06%)181.76 (+54.85%)182.20 (+57.75%)123.70 (+68.07%)40.71 (+26.92%)
1531781 — 2026-01-23 18:49:241.34 (n/a)0.89 (n/a)0.85 (n/a)0.60 (n/a)0.27 (n/a)163.50 (n/a)117.38 (n/a)115.50 (n/a)73.60 (n/a)32.08 (n/a)

rope_8c_32rows_512cols_8arows_0m

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.56 (+2.22%)0.44 (-15.04%)0.43 (-18.14%)0.31 (-35.43%)0.12 (+342.48%)236.70 (+54.81%)178.62 (+24.63%)170.70 (+22.19%)132.60 (-2.14%)48.54 (+550.30%)
1531781 — 2026-01-23 18:49:240.54 (n/a)0.52 (n/a)0.53 (n/a)0.48 (n/a)0.03 (n/a)152.90 (n/a)143.32 (n/a)139.70 (n/a)135.50 (n/a)7.46 (n/a)

sigmoid_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.09 (-21.48%)0.06 (-35.34%)0.07 (-43.17%)0.04 (-24.92%)0.02 (-31.40%)204.10 (+33.22%)138.18 (+51.98%)120.60 (+75.80%)87.00 (+27.38%)44.81 (+21.76%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.10 (n/a)0.12 (n/a)0.05 (n/a)0.03 (n/a)153.20 (n/a)90.92 (n/a)68.60 (n/a)68.30 (n/a)36.80 (n/a)

sigmoid_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.09 (-28.38%)0.06 (-40.97%)0.06 (-52.67%)0.04 (-32.25%)0.02 (-40.40%)207.90 (+47.66%)148.86 (+65.69%)145.70 (+111.47%)95.10 (+39.65%)40.20 (+25.47%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.10 (n/a)0.12 (n/a)0.06 (n/a)0.03 (n/a)140.80 (n/a)89.84 (n/a)68.90 (n/a)68.10 (n/a)32.04 (n/a)

sigmoid_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.07 (-23.57%)0.05 (-38.76%)0.04 (-52.00%)0.04 (-18.69%)0.01 (-32.98%)208.40 (+23.02%)170.38 (+59.86%)182.20 (+108.47%)109.90 (+30.83%)37.56 (+3.28%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.08 (n/a)0.09 (n/a)0.05 (n/a)0.02 (n/a)169.40 (n/a)106.58 (n/a)87.40 (n/a)84.00 (n/a)36.37 (n/a)

sigmoid_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.06 (-49.43%)0.05 (-54.03%)0.04 (-65.53%)0.04 (-28.13%)0.01 (-64.85%)218.90 (+39.16%)182.16 (+102.99%)199.30 (+190.10%)130.40 (+97.88%)37.23 (-4.16%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.10 (n/a)0.12 (n/a)0.05 (n/a)0.03 (n/a)157.30 (n/a)89.74 (n/a)68.70 (n/a)65.90 (n/a)38.85 (n/a)

silu_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.07 (-34.12%)0.05 (-39.70%)0.05 (-45.50%)0.04 (-36.12%)0.01 (-41.48%)224.40 (+56.59%)172.38 (+64.36%)150.90 (+83.58%)122.30 (+51.74%)48.14 (+49.33%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.08 (n/a)0.10 (n/a)0.06 (n/a)0.02 (n/a)143.30 (n/a)104.88 (n/a)82.20 (n/a)80.60 (n/a)32.24 (n/a)

silu_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.07 (-30.32%)0.05 (-43.67%)0.05 (-53.57%)0.03 (-49.16%)0.02 (-27.92%)273.20 (+96.69%)187.30 (+81.67%)179.30 (+115.50%)117.10 (+43.50%)58.45 (+106.51%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.08 (n/a)0.10 (n/a)0.06 (n/a)0.02 (n/a)138.90 (n/a)103.10 (n/a)83.20 (n/a)81.60 (n/a)28.30 (n/a)

silu_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.08 (-36.24%)0.06 (-47.34%)0.06 (-52.05%)0.04 (-37.10%)0.02 (-37.98%)218.40 (+58.95%)157.60 (+89.06%)145.90 (+108.43%)106.80 (+56.83%)46.62 (+54.00%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.11 (n/a)0.12 (n/a)0.06 (n/a)0.03 (n/a)137.40 (n/a)83.36 (n/a)70.00 (n/a)68.10 (n/a)30.27 (n/a)

silu_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.05 (-59.21%)0.04 (-53.76%)0.05 (-51.47%)0.04 (-48.84%)0.00 (-74.77%)214.60 (+95.45%)190.12 (+113.14%)181.20 (+106.14%)174.30 (+145.15%)17.50 (+20.62%)
1531781 — 2026-01-23 18:49:240.12 (n/a)0.09 (n/a)0.09 (n/a)0.07 (n/a)0.02 (n/a)109.80 (n/a)89.20 (n/a)87.90 (n/a)71.10 (n/a)14.50 (n/a)

softmax_1_cols_2_channels_4096_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.21 (+20.19%)0.16 (+13.84%)0.16 (+8.19%)0.12 (+26.61%)0.04 (+17.18%)135.50 (-20.99%)104.68 (-12.65%)102.70 (-7.56%)76.50 (-16.85%)22.55 (-26.21%)
0f0da0b — 2026-01-14 16:31:000.18 (n/a)0.14 (n/a)0.15 (n/a)0.10 (n/a)0.03 (n/a)171.50 (n/a)119.84 (n/a)111.10 (n/a)92.00 (n/a)30.56 (n/a)

softmax_2_cols_2_channels_32768_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.92 (n/a)0.77 (n/a)0.84 (n/a)0.59 (n/a)0.16 (n/a)223.90 (n/a)178.12 (n/a)156.90 (n/a)142.80 (n/a)40.41 (n/a)

softmax_2_cols_2_channels_32768_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:361.20 (n/a)0.89 (n/a)0.80 (n/a)0.62 (n/a)0.23 (n/a)212.00 (n/a)154.70 (n/a)163.10 (n/a)109.70 (n/a)39.84 (n/a)

softmax_2_cols_2_channels_4096_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.22 (+40.26%)0.18 (+43.33%)0.20 (+64.08%)0.13 (+24.34%)0.04 (+68.98%)127.80 (-19.62%)94.74 (-28.95%)83.40 (-39.08%)73.90 (-28.67%)24.46 (-5.31%)
0f0da0b — 2026-01-14 16:31:000.16 (n/a)0.13 (n/a)0.12 (n/a)0.10 (n/a)0.03 (n/a)159.00 (n/a)133.34 (n/a)136.90 (n/a)103.60 (n/a)25.83 (n/a)

softmax_2_cols_2_channels_4096_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.24 (+85.21%)0.17 (+69.97%)0.15 (+54.16%)0.11 (+37.21%)0.05 (+203.82%)148.10 (-27.12%)102.78 (-37.88%)106.70 (-35.10%)69.60 (-45.96%)31.81 (+15.98%)
0f0da0b — 2026-01-14 16:31:000.13 (n/a)0.10 (n/a)0.10 (n/a)0.08 (n/a)0.02 (n/a)203.20 (n/a)165.46 (n/a)164.40 (n/a)128.80 (n/a)27.43 (n/a)

softmax_4_cols_4_channels_32768_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.61 (n/a)0.59 (n/a)0.61 (n/a)0.53 (n/a)0.04 (n/a)245.30 (n/a)223.14 (n/a)214.10 (n/a)213.10 (n/a)14.06 (n/a)

swiglu

No metrics available.

swiglu_decode_1x2048x2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:36n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)
130b6ea — 2025-12-05 21:15:490.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)5187.20 (n/a)4720.84 (n/a)4750.90 (n/a)4015.10 (n/a)390.23 (n/a)

swiglu_decode_1x2048x2048_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.00 (+28.14%)0.00 (+20.30%)0.00 (+10.00%)0.00 (+11.66%)0.00 (+124.73%)5111.40 (-10.44%)4324.88 (-14.91%)4736.10 (-9.09%)3410.70 (-21.96%)843.97 (+53.51%)
0f0da0b — 2026-01-14 16:31:000.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)5707.20 (n/a)5083.00 (n/a)5209.60 (n/a)4370.50 (n/a)549.78 (n/a)

swiglu_prefill_256x2048x2048

No metrics available.

tanh_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.05 (-46.26%)0.05 (-38.80%)0.05 (-29.94%)0.04 (-25.32%)0.00 (-78.80%)188.60 (+33.95%)172.58 (+58.13%)165.50 (+42.80%)159.30 (+86.10%)13.18 (-43.97%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.08 (n/a)0.07 (n/a)0.06 (n/a)0.02 (n/a)140.80 (n/a)109.14 (n/a)115.90 (n/a)85.60 (n/a)23.52 (n/a)

tanh_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.06 (-46.34%)0.05 (-45.76%)0.05 (-44.71%)0.05 (-41.58%)0.00 (-53.10%)176.90 (+71.25%)161.90 (+83.94%)158.80 (+80.87%)145.00 (+86.38%)14.54 (+50.84%)
1531781 — 2026-01-23 18:49:240.11 (n/a)0.09 (n/a)0.09 (n/a)0.08 (n/a)0.01 (n/a)103.30 (n/a)88.02 (n/a)87.80 (n/a)77.80 (n/a)9.64 (n/a)

tanh_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.06 (-36.61%)0.05 (-43.80%)0.05 (-48.65%)0.04 (-36.29%)0.01 (-45.17%)190.50 (+56.92%)165.32 (+77.00%)168.10 (+94.79%)133.50 (+57.80%)20.77 (+31.42%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.09 (n/a)0.09 (n/a)0.07 (n/a)0.01 (n/a)121.40 (n/a)93.40 (n/a)86.30 (n/a)84.60 (n/a)15.80 (n/a)

tanh_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.05 (-52.66%)0.05 (-47.71%)0.04 (-57.81%)0.04 (-20.22%)0.01 (-78.31%)205.30 (+25.34%)179.78 (+74.71%)190.50 (+136.94%)151.40 (+111.16%)23.19 (-42.01%)
1531781 — 2026-01-23 18:49:240.11 (n/a)0.09 (n/a)0.10 (n/a)0.05 (n/a)0.03 (n/a)163.80 (n/a)102.90 (n/a)80.40 (n/a)71.70 (n/a)39.99 (n/a)

transpose_2048_M_64_N_1_cols_1_channels_64_m_64_n_8_s

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:363.65 (-6.96%)2.97 (-19.36%)3.00 (-17.85%)2.20 (-36.26%)0.52 (+240.76%)238.20 (+56.92%)181.42 (+27.13%)174.50 (+21.69%)143.50 (+7.49%)34.83 (+490.55%)
130b6ea — 2025-12-05 21:15:493.93 (n/a)3.68 (n/a)3.66 (n/a)3.45 (n/a)0.15 (n/a)151.80 (n/a)142.70 (n/a)143.40 (n/a)133.50 (n/a)5.90 (n/a)

transpose_2048_M_64_N_1_cols_1_channels_64_m_64_n_8_s0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:244.95 (+35.28%)4.34 (+33.32%)4.37 (+25.58%)3.74 (+35.73%)0.50 (+11.62%)140.30 (-26.35%)121.98 (-25.39%)119.90 (-20.33%)106.00 (-26.08%)14.15 (-39.67%)
0f0da0b — 2026-01-14 16:31:003.66 (n/a)3.26 (n/a)3.48 (n/a)2.75 (n/a)0.45 (n/a)190.50 (n/a)163.48 (n/a)150.50 (n/a)143.40 (n/a)23.46 (n/a)

transpose_2048_M_64_N_1_cols_2_channels_64_m_64_n_8_s

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:363.84 (-24.88%)3.52 (-8.71%)3.75 (+10.15%)2.90 (-11.69%)0.40 (-43.11%)180.70 (+13.22%)150.72 (+7.52%)140.00 (-9.21%)136.60 (+33.14%)18.92 (-16.63%)
130b6ea — 2025-12-05 21:15:495.11 (n/a)3.85 (n/a)3.40 (n/a)3.29 (n/a)0.71 (n/a)159.60 (n/a)140.18 (n/a)154.20 (n/a)102.60 (n/a)22.70 (n/a)

transpose_2048_M_64_N_1_cols_2_channels_64_m_64_n_8_s0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:246.07 (+44.96%)4.99 (+45.77%)5.09 (+52.05%)4.09 (+35.80%)0.74 (+53.13%)128.10 (-26.34%)106.86 (-31.23%)103.00 (-34.27%)86.30 (-31.02%)15.67 (-22.94%)
0f0da0b — 2026-01-14 16:31:004.19 (n/a)3.42 (n/a)3.35 (n/a)3.01 (n/a)0.48 (n/a)173.90 (n/a)155.38 (n/a)156.70 (n/a)125.10 (n/a)20.34 (n/a)

weighted_rms_norm_1_cols_2_channels_2048_weights_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.10 (-7.83%)0.08 (-3.92%)0.08 (-11.90%)0.07 (+40.53%)0.01 (-48.04%)165.50 (+6.77%)148.78 (+48.90%)152.60 (+70.31%)124.50 (+62.75%)17.93 (-42.84%)
1531781 — 2026-01-23 18:49:240.11 (n/a)0.09 (n/a)0.09 (n/a)0.05 (n/a)0.02 (n/a)155.00 (n/a)99.92 (n/a)89.60 (n/a)76.50 (n/a)31.36 (n/a)

weighted_rms_norm_2_cols_2_channels_2048_weights_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.08 (-20.64%)0.07 (-22.49%)0.07 (-21.52%)0.04 (-15.66%)0.01 (-27.37%)247.00 (+48.17%)163.12 (+58.80%)140.90 (+59.21%)130.90 (+57.52%)48.25 (+34.53%)
1531781 — 2026-01-23 18:49:240.10 (n/a)0.09 (n/a)0.09 (n/a)0.05 (n/a)0.02 (n/a)166.70 (n/a)102.72 (n/a)88.50 (n/a)83.10 (n/a)35.87 (n/a)

weighted_rms_norm_4_cols_2_channels_2048_weights_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.07 (-22.43%)0.06 (-30.21%)0.06 (-34.27%)0.05 (-26.78%)0.01 (-3.84%)182.90 (+53.70%)156.76 (+62.48%)153.80 (+71.08%)129.50 (+45.02%)24.73 (+93.13%)
1531781 — 2026-01-23 18:49:240.09 (n/a)0.09 (n/a)0.09 (n/a)0.07 (n/a)0.01 (n/a)119.00 (n/a)96.48 (n/a)89.90 (n/a)89.30 (n/a)12.81 (n/a)

weighted_rms_norm_8_cols_2_channels_2048_weights_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
9369dae — 2026-02-09 20:32:360.05 (-54.93%)0.04 (-55.29%)0.04 (-58.09%)0.04 (-37.87%)0.00 (-76.45%)232.20 (+70.99%)205.40 (+128.43%)202.90 (+153.62%)175.20 (+135.80%)22.07 (-14.31%)
1531781 — 2026-01-23 18:49:240.11 (n/a)0.10 (n/a)0.10 (n/a)0.06 (n/a)0.02 (n/a)135.80 (n/a)89.92 (n/a)80.00 (n/a)74.30 (n/a)25.76 (n/a)

@github-actions
Copy link
Contributor

github-actions bot commented Feb 9, 2026

📊 Test Results for Small Benchmark/Test Suite

82cf4ab (2026_02_09_20_49_27)

IRONCLAD

Tested on 2026_02_09_20_49_27 at commit 82cf4ab.

Test Checks Latency (mean)Bandwidth (mean)GFLOP/s (mean)
M_128-K_128-num_aie_columns_1-tile_size_input_32-tile_size_output_128 ✅ 5/5 n/a 0.19 n/a
M_1792-K_896-N_1152-num_aie_columns_8-b_col_maj_False-c_col_maj_True-m_64-k_32-n_48-trace_size_0-partition_N_1 ✅ 5/5 2244.96 4.26 n/a
M_192-K_384-N_64-num_aie_columns_4-b_col_maj_False-c_col_maj_False-m_48-k_96-n_16-trace_size_0-partition_N_1 ✅ 5/5 259.82 1.00 n/a
M_192-K_384-N_64-num_aie_columns_4-b_col_maj_True-c_col_maj_True-m_48-k_96-n_16-trace_size_0-partition_N_1 ✅ 5/5 229.94 0.96 n/a
M_2048-K_2048-N_2048-num_aie_columns_1-b_col_maj_False-c_col_maj_False-m_64-k_64-n_64-trace_size_0-partition_N_1 ✅ 5/5 48621.56 0.52 n/a
M_2048-K_2048-N_2048-num_aie_columns_2-b_col_maj_True-c_col_maj_False-m_64-k_64-n_64-trace_size_0-partition_N_1 ✅ 5/5 27786.84 0.91 n/a
M_2048-K_2048-N_2048-num_aie_columns_8-b_col_maj_True-c_col_maj_True-m_64-k_64-n_64-trace_size_0-partition_N_1 ✅ 5/5 7792.70 3.23 n/a
M_2048-K_8192-num_aie_columns_1-tile_size_input_1-tile_size_output_2048 ✅ 5/5 n/a 12.64 n/a
M_2048-K_8192-num_aie_columns_2-tile_size_input_1-tile_size_output_1024 ✅ 5/5 n/a 24.10 n/a
M_2048-K_8192-num_aie_columns_4-tile_size_input_1-tile_size_output_512 ✅ 5/5 n/a 40.12 n/a
M_2048-K_8192-num_aie_columns_8-tile_size_input_1-tile_size_output_256 ✅ 5/5 n/a 42.63 n/a
M_2048-N_64-aie_columns_1-channels_1-m_64-n_64-s_8 ✅ 5/5 218.96 2.65 n/a
M_2048-N_64-aie_columns_1-channels_2-m_64-n_64-s_8 ✅ 5/5 168.24 3.22 n/a
M_384-K_1536-N_1792-num_aie_columns_4-b_col_maj_True-c_col_maj_False-m_32-k_48-n_64-trace_size_0-partition_N_1 ✅ 5/5 2197.54 3.73 n/a
M_8192-K_2048-num_aie_columns_1-tile_size_input_4-tile_size_output_1024 ✅ 5/5 n/a 13.12 n/a
M_8192-K_2048-num_aie_columns_2-tile_size_input_4-tile_size_output_1024 ✅ 5/5 n/a 24.07 n/a
M_8192-K_2048-num_aie_columns_4-tile_size_input_4-tile_size_output_1024 ✅ 5/5 n/a 39.40 n/a
M_8192-K_2048-num_aie_columns_8-tile_size_input_4-tile_size_output_1024 ✅ 5/5 n/a 42.35 n/a
M_896-K_1792-N_640-num_aie_columns_8-b_col_maj_False-c_col_maj_True-m_32-k_64-n_80-trace_size_0-partition_N_1 ✅ 5/5 1220.58 5.59 n/a
embedding_dim_2048-hidden_dim_2048 ✅ 5/5 n/a n/a n/a
input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048 ✅ 30/30 162.77 0.05 n/a
input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048-group_size_32 ✅ 5/5 176.28 0.03 n/a
input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048-weighted_False ✅ 5/5 164.50 0.05 n/a
input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024 ✅ 10/10 157.19 0.05 n/a
input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024-group_size_32 ✅ 5/5 157.18 0.03 n/a
input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024-weighted_False ✅ 5/5 168.68 0.05 n/a
input_length_2048-num_aie_columns_1-num_channels_2-tile_size_2048 ✅ 10/10 159.43 0.08 n/a
input_length_2048-num_aie_columns_1-num_channels_2-tile_size_2048-scalar_factor_3.0 ✅ 5/5 151.40 0.09 n/a
input_length_2048-num_aie_columns_1-num_channels_2-tile_size_2048-weighted_True ✅ 5/5 185.10 0.07 n/a
input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024 ✅ 30/30 155.12 0.06 n/a
input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024-group_size_32 ✅ 5/5 137.42 0.04 n/a
input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024-weighted_False ✅ 5/5 164.18 0.05 n/a
input_length_2048-num_aie_columns_2-num_channels_2-tile_size_1024 ✅ 10/10 167.48 0.08 n/a
input_length_2048-num_aie_columns_2-num_channels_2-tile_size_1024-scalar_factor_3.0 ✅ 5/5 141.44 0.09 n/a
input_length_2048-num_aie_columns_2-num_channels_2-tile_size_1024-weighted_True ✅ 5/5 208.82 0.05 n/a
input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512 ✅ 10/10 183.86 0.05 n/a
input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512-group_size_32 ✅ 5/5 168.68 0.03 n/a
input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512-weighted_False ✅ 5/5 166.40 0.05 n/a
input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512 ✅ 30/30 161.01 0.05 n/a
input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512-group_size_32 ✅ 5/5 162.64 0.03 n/a
input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512-weighted_False ✅ 5/5 204.18 0.04 n/a
input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256 ✅ 10/10 165.15 0.05 n/a
input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256-group_size_32 ✅ 5/5 192.64 0.03 n/a
input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256-weighted_False ✅ 5/5 194.78 0.04 n/a
input_length_2048-num_aie_columns_4-num_channels_2-tile_size_512 ✅ 10/10 175.70 0.07 n/a
input_length_2048-num_aie_columns_4-num_channels_2-tile_size_512-scalar_factor_3.0 ✅ 5/5 155.18 0.08 n/a
input_length_2048-num_aie_columns_4-num_channels_2-tile_size_512-weighted_True ✅ 5/5 183.10 0.05 n/a
input_length_2048-num_aie_columns_8-num_channels_1-tile_size_256 ✅ 30/30 187.23 0.05 n/a
input_length_2048-num_aie_columns_8-num_channels_1-tile_size_256-group_size_32 ✅ 5/5 173.28 0.03 n/a
input_length_2048-num_aie_columns_8-num_channels_1-tile_size_256-weighted_False ✅ 5/5 191.00 0.04 n/a
input_length_2048-num_aie_columns_8-num_channels_2-tile_size_128 ✅ 10/10 223.23 0.04 n/a
input_length_2048-num_aie_columns_8-num_channels_2-tile_size_128-group_size_32 ✅ 5/5 192.52 0.03 n/a
input_length_2048-num_aie_columns_8-num_channels_2-tile_size_128-weighted_False ✅ 5/5 280.10 0.03 n/a
input_length_2048-num_aie_columns_8-num_channels_2-tile_size_256 ✅ 10/10 187.51 0.07 n/a
input_length_2048-num_aie_columns_8-num_channels_2-tile_size_256-scalar_factor_3.0 ✅ 5/5 154.94 0.08 n/a
input_length_2048-num_aie_columns_8-num_channels_2-tile_size_256-weighted_True ✅ 5/5 249.82 0.04 n/a
input_length_2048-num_cores_1-num_channels_1-bypass_False-tile_size_2048 ✅ 5/5 176.58 0.05 n/a
input_length_2048-num_cores_16-num_channels_2-bypass_False-tile_size_128 ✅ 5/5 212.50 0.04 n/a
input_length_2048-num_cores_2-num_channels_1-bypass_False-tile_size_1024 ✅ 5/5 159.70 0.05 n/a
input_length_2048-num_cores_2-num_channels_2-bypass_False-tile_size_1024 ✅ 5/5 149.42 0.06 n/a
input_length_2048-num_cores_4-num_channels_1-bypass_False-tile_size_512 ✅ 5/5 184.16 0.05 n/a
input_length_2048-num_cores_4-num_channels_2-bypass_False-tile_size_512 ✅ 5/5 150.00 0.06 n/a
input_length_2048-num_cores_8-num_channels_1-bypass_False-tile_size_256 ✅ 5/5 177.92 0.05 n/a
input_length_2048-num_cores_8-num_channels_2-bypass_False-tile_size_256 ✅ 5/5 178.72 0.05 n/a
input_length_32768-num_aie_columns_2-num_channels_2-tile_size_1024 ✅ 5/5 128.58 1.03 n/a
input_length_32768-num_aie_columns_2-num_channels_2-tile_size_512 ✅ 5/5 160.80 0.86 n/a
input_length_32768-num_aie_columns_4-num_channels_4-tile_size_2048 ✅ 5/5 201.22 0.77 n/a
rows_32-cols_512-angle_rows_32-aie_columns_1-method_type_0 ✅ 5/5 160.66 0.64 n/a
rows_32-cols_512-angle_rows_32-aie_columns_2-method_type_0 ✅ 5/5 166.24 0.63 n/a
rows_32-cols_512-angle_rows_32-aie_columns_8-method_type_0 ✅ 5/5 176.82 0.58 n/a
rows_32-cols_512-angle_rows_8-aie_columns_1-method_type_0 ✅ 5/5 155.70 0.49 n/a
rows_32-cols_512-angle_rows_8-aie_columns_2-method_type_0 ✅ 5/5 168.84 0.46 n/a
rows_32-cols_512-angle_rows_8-aie_columns_8-method_type_0 ✅ 5/5 244.34 0.32 n/a
seq_len_16384-dim_64-num_heads_1-num_pipelines_8 ✅ 5/5 47504.22 0.18 n/a
seq_len_256-embedding_dim_2048-hidden_dim_2048-prio_accuracy_False ❌ 0/5 n/a n/a n/a
📈 Trends (vs main branch) for Small Benchmark/Test Suite

82cf4ab (2026_02_09_20_49_27)

IRONCLAD Trends

M_128-K_128-num_aie_columns_1-tile_size_input_32-tile_size_output_128

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
82cf4ab — 2026-02-09 20:43:010.22 (n/a)0.19 (n/a)0.19 (n/a)0.14 (n/a)0.03 (n/a)0.22 (n/a)0.18 (n/a)0.19 (n/a)0.14 (n/a)0.03 (n/a)

M_1792-K_896-N_1152-num_aie_columns_8-b_col_maj_False-c_col_maj_True-m_64-k_32-n_48-trace_size_0-partition_N_1

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
82cf4ab — 2026-02-09 20:43:014.99 (n/a)4.26 (n/a)4.10 (n/a)3.43 (n/a)0.62 (n/a)2743.20 (n/a)2244.96 (n/a)2292.20 (n/a)1884.00 (n/a)338.63 (n/a)1963.59 (n/a)1677.16 (n/a)1613.93 (n/a)1348.54 (n/a)244.40 (n/a)

M_192-K_384-N_64-num_aie_columns_4-b_col_maj_False-c_col_maj_False-m_48-k_96-n_16-trace_size_0-partition_N_1

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
82cf4ab — 2026-02-09 20:43:011.68 (n/a)1.00 (n/a)1.01 (n/a)0.52 (n/a)0.45 (n/a)422.20 (n/a)259.82 (n/a)218.60 (n/a)131.40 (n/a)116.20 (n/a)71.84 (n/a)42.84 (n/a)43.17 (n/a)22.35 (n/a)19.37 (n/a)

M_192-K_384-N_64-num_aie_columns_4-b_col_maj_True-c_col_maj_True-m_48-k_96-n_16-trace_size_0-partition_N_1

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
82cf4ab — 2026-02-09 20:43:010.99 (n/a)0.96 (n/a)0.97 (n/a)0.91 (n/a)0.03 (n/a)243.20 (n/a)229.94 (n/a)227.50 (n/a)223.40 (n/a)7.66 (n/a)42.24 (n/a)41.08 (n/a)41.49 (n/a)38.81 (n/a)1.32 (n/a)

M_2048-K_2048-N_2048-num_aie_columns_1-b_col_maj_False-c_col_maj_False-m_64-k_64-n_64-trace_size_0-partition_N_1

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
82cf4ab — 2026-02-09 20:43:010.52 (n/a)0.52 (n/a)0.52 (n/a)0.52 (n/a)0.00 (n/a)48819.80 (n/a)48621.56 (n/a)48587.80 (n/a)48513.40 (n/a)116.20 (n/a)354.13 (n/a)353.34 (n/a)353.58 (n/a)351.90 (n/a)0.84 (n/a)

M_2048-K_2048-N_2048-num_aie_columns_2-b_col_maj_True-c_col_maj_False-m_64-k_64-n_64-trace_size_0-partition_N_1

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
82cf4ab — 2026-02-09 20:43:010.91 (n/a)0.91 (n/a)0.91 (n/a)0.90 (n/a)0.00 (n/a)27867.80 (n/a)27786.84 (n/a)27779.60 (n/a)27712.60 (n/a)57.26 (n/a)619.93 (n/a)618.28 (n/a)618.44 (n/a)616.48 (n/a)1.27 (n/a)

M_2048-K_2048-N_2048-num_aie_columns_8-b_col_maj_True-c_col_maj_True-m_64-k_64-n_64-trace_size_0-partition_N_1

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
82cf4ab — 2026-02-09 20:43:013.44 (n/a)3.23 (n/a)3.19 (n/a)3.10 (n/a)0.14 (n/a)8106.70 (n/a)7792.70 (n/a)7882.80 (n/a)7325.60 (n/a)330.76 (n/a)2345.17 (n/a)2207.85 (n/a)2179.40 (n/a)2119.22 (n/a)95.48 (n/a)

M_2048-K_8192-num_aie_columns_1-tile_size_input_1-tile_size_output_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
82cf4ab — 2026-02-09 20:43:0113.32 (n/a)12.64 (n/a)12.51 (n/a)12.10 (n/a)0.55 (n/a)13.31 (n/a)12.63 (n/a)12.50 (n/a)12.09 (n/a)0.55 (n/a)

M_2048-K_8192-num_aie_columns_2-tile_size_input_1-tile_size_output_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
82cf4ab — 2026-02-09 20:43:0124.80 (n/a)24.10 (n/a)24.29 (n/a)22.75 (n/a)0.81 (n/a)24.79 (n/a)24.09 (n/a)24.28 (n/a)22.73 (n/a)0.81 (n/a)

M_2048-K_8192-num_aie_columns_4-tile_size_input_1-tile_size_output_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
82cf4ab — 2026-02-09 20:43:0142.06 (n/a)40.12 (n/a)39.90 (n/a)38.70 (n/a)1.23 (n/a)42.04 (n/a)40.09 (n/a)39.88 (n/a)38.67 (n/a)1.23 (n/a)

M_2048-K_8192-num_aie_columns_8-tile_size_input_1-tile_size_output_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
82cf4ab — 2026-02-09 20:43:0144.17 (n/a)42.63 (n/a)42.51 (n/a)41.04 (n/a)1.15 (n/a)44.15 (n/a)42.60 (n/a)42.48 (n/a)41.02 (n/a)1.15 (n/a)

M_2048-N_64-aie_columns_1-channels_1-m_64-n_64-s_8

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
82cf4ab — 2026-02-09 20:43:013.75 (n/a)2.65 (n/a)2.35 (n/a)1.76 (n/a)0.95 (n/a)297.80 (n/a)218.96 (n/a)222.90 (n/a)140.00 (n/a)74.50 (n/a)

M_2048-N_64-aie_columns_1-channels_2-m_64-n_64-s_8

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
82cf4ab — 2026-02-09 20:43:014.01 (n/a)3.22 (n/a)3.30 (n/a)2.32 (n/a)0.61 (n/a)226.30 (n/a)168.24 (n/a)158.80 (n/a)130.70 (n/a)35.56 (n/a)

M_384-K_1536-N_1792-num_aie_columns_4-b_col_maj_True-c_col_maj_False-m_32-k_48-n_64-trace_size_0-partition_N_1

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
82cf4ab — 2026-02-09 20:43:014.25 (n/a)3.73 (n/a)3.74 (n/a)2.93 (n/a)0.53 (n/a)2749.40 (n/a)2197.54 (n/a)2153.40 (n/a)1898.30 (n/a)341.42 (n/a)1113.60 (n/a)979.01 (n/a)981.67 (n/a)768.86 (n/a)138.16 (n/a)

M_8192-K_2048-num_aie_columns_1-tile_size_input_4-tile_size_output_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
82cf4ab — 2026-02-09 20:43:0113.23 (n/a)13.12 (n/a)13.16 (n/a)12.98 (n/a)0.10 (n/a)13.22 (n/a)13.11 (n/a)13.15 (n/a)12.97 (n/a)0.10 (n/a)

M_8192-K_2048-num_aie_columns_2-tile_size_input_4-tile_size_output_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
82cf4ab — 2026-02-09 20:43:0124.82 (n/a)24.07 (n/a)24.19 (n/a)22.88 (n/a)0.72 (n/a)24.80 (n/a)24.05 (n/a)24.18 (n/a)22.86 (n/a)0.72 (n/a)

M_8192-K_2048-num_aie_columns_4-tile_size_input_4-tile_size_output_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
82cf4ab — 2026-02-09 20:43:0142.28 (n/a)39.40 (n/a)38.67 (n/a)38.00 (n/a)1.78 (n/a)42.25 (n/a)39.38 (n/a)38.64 (n/a)37.98 (n/a)1.78 (n/a)

M_8192-K_2048-num_aie_columns_8-tile_size_input_4-tile_size_output_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
82cf4ab — 2026-02-09 20:43:0146.01 (n/a)42.35 (n/a)41.50 (n/a)40.34 (n/a)2.20 (n/a)45.98 (n/a)42.33 (n/a)41.48 (n/a)40.31 (n/a)2.20 (n/a)

M_896-K_1792-N_640-num_aie_columns_8-b_col_maj_False-c_col_maj_True-m_32-k_64-n_80-trace_size_0-partition_N_1

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
82cf4ab — 2026-02-09 20:43:016.63 (n/a)5.59 (n/a)5.90 (n/a)4.49 (n/a)0.97 (n/a)1483.00 (n/a)1220.58 (n/a)1128.00 (n/a)1003.30 (n/a)220.77 (n/a)2048.52 (n/a)1727.06 (n/a)1822.06 (n/a)1385.88 (n/a)299.63 (n/a)

axpy_1_cols_2_channels_2048_tile_2048_3.0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.13 (-1.96%)0.11 (+2.52%)0.10 (+1.86%)0.08 (-1.91%)0.02 (-1.72%)162.40 (+1.95%)120.96 (-2.36%)117.00 (-1.85%)94.40 (+2.05%)23.11 (+4.09%)
0a6c11c — 2025-12-04 00:39:100.13 (n/a)0.10 (n/a)0.10 (n/a)0.08 (n/a)0.02 (n/a)159.30 (n/a)123.88 (n/a)119.20 (n/a)92.50 (n/a)22.21 (n/a)

axpy_1_cols_2_channels_2048_tile_2048_3.0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.16 (+80.10%)0.13 (+54.74%)0.11 (+34.32%)0.10 (+39.85%)0.03 (+432.10%)118.80 (-28.48%)100.30 (-33.22%)110.40 (-25.56%)78.00 (-44.48%)20.49 (+105.98%)
0f0da0b — 2026-01-14 16:31:000.09 (n/a)0.08 (n/a)0.08 (n/a)0.07 (n/a)0.01 (n/a)166.10 (n/a)150.20 (n/a)148.30 (n/a)140.50 (n/a)9.95 (n/a)

axpy_2_cols_2_channels_2048_tile_1024_3.0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.11 (-26.77%)0.09 (-9.09%)0.09 (-10.81%)0.08 (+8.25%)0.01 (-55.42%)152.00 (-7.60%)135.46 (+6.28%)142.80 (+12.18%)115.00 (+36.58%)15.03 (-41.36%)
0a6c11c — 2025-12-04 00:39:100.15 (n/a)0.10 (n/a)0.10 (n/a)0.07 (n/a)0.02 (n/a)164.50 (n/a)127.46 (n/a)127.30 (n/a)84.20 (n/a)25.64 (n/a)

axpy_2_cols_2_channels_2048_tile_1024_3.0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.14 (+54.25%)0.12 (+45.65%)0.14 (+61.31%)0.08 (-2.79%)0.03 (+557.87%)156.90 (+2.89%)104.46 (-28.09%)90.30 (-37.98%)87.90 (-35.22%)29.52 (+351.33%)
0f0da0b — 2026-01-14 16:31:000.09 (n/a)0.08 (n/a)0.08 (n/a)0.08 (n/a)0.00 (n/a)152.50 (n/a)145.26 (n/a)145.60 (n/a)135.70 (n/a)6.54 (n/a)

axpy_4_cols_2_channels_2048_tile_512_3.0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.12 (-10.21%)0.10 (-8.41%)0.10 (-6.20%)0.07 (+3.52%)0.02 (-23.48%)166.70 (-3.36%)128.78 (+7.33%)121.80 (+6.65%)103.90 (+11.36%)22.95 (-18.82%)
0a6c11c — 2025-12-04 00:39:100.13 (n/a)0.11 (n/a)0.11 (n/a)0.07 (n/a)0.02 (n/a)172.50 (n/a)119.98 (n/a)114.20 (n/a)93.30 (n/a)28.27 (n/a)

axpy_4_cols_2_channels_2048_tile_512_3.0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.16 (+65.19%)0.14 (+57.55%)0.15 (+63.48%)0.09 (+24.56%)0.03 (+220.28%)134.10 (-19.70%)92.44 (-34.51%)83.40 (-38.81%)78.60 (-39.49%)23.38 (+58.01%)
0f0da0b — 2026-01-14 16:31:000.09 (n/a)0.09 (n/a)0.09 (n/a)0.07 (n/a)0.01 (n/a)167.00 (n/a)141.16 (n/a)136.30 (n/a)129.90 (n/a)14.80 (n/a)

axpy_8_cols_2_channels_2048_tile_256_3.0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.10 (-16.19%)0.08 (-10.40%)0.09 (-2.09%)0.07 (-17.38%)0.01 (-10.61%)184.40 (+21.00%)153.18 (+11.99%)144.10 (+2.20%)127.10 (+19.34%)21.26 (+34.76%)
0a6c11c — 2025-12-04 00:39:100.12 (n/a)0.09 (n/a)0.09 (n/a)0.08 (n/a)0.01 (n/a)152.40 (n/a)136.78 (n/a)141.00 (n/a)106.50 (n/a)15.78 (n/a)

axpy_8_cols_2_channels_2048_tile_256_3.0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.17 (+67.03%)0.14 (+66.86%)0.15 (+90.72%)0.08 (+20.53%)0.03 (+162.97%)146.80 (-17.06%)92.24 (-37.11%)80.00 (-47.54%)72.30 (-40.15%)31.07 (+36.92%)
0f0da0b — 2026-01-14 16:31:000.10 (n/a)0.09 (n/a)0.08 (n/a)0.07 (n/a)0.01 (n/a)177.00 (n/a)146.68 (n/a)152.50 (n/a)120.80 (n/a)22.69 (n/a)

dequant_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.05 (-18.83%)0.04 (-5.21%)0.04 (-2.69%)0.03 (+37.10%)0.00 (-53.98%)160.40 (-27.06%)141.12 (-0.91%)136.40 (+2.79%)114.30 (+23.17%)17.37 (-59.06%)
0a6c11c — 2025-12-04 00:39:100.06 (n/a)0.04 (n/a)0.04 (n/a)0.02 (n/a)0.01 (n/a)219.90 (n/a)142.42 (n/a)132.70 (n/a)92.80 (n/a)42.43 (n/a)

dequant_1_cols_1_channels_2048_tile_2048_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.06 (+13.11%)0.05 (+1.55%)0.04 (-0.75%)0.04 (+7.05%)0.01 (+25.85%)125.50 (-6.55%)115.08 (-1.12%)121.60 (+0.75%)88.10 (-11.63%)15.40 (+4.09%)
0f0da0b — 2026-01-14 16:31:000.05 (n/a)0.05 (n/a)0.04 (n/a)0.04 (n/a)0.01 (n/a)134.30 (n/a)116.38 (n/a)120.70 (n/a)99.70 (n/a)14.79 (n/a)

dequant_1_cols_2_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.06 (-13.28%)0.05 (-8.93%)0.06 (-13.59%)0.04 (+4.63%)0.01 (-39.10%)133.60 (-4.43%)106.08 (+5.85%)95.10 (+15.69%)87.90 (+15.35%)17.90 (-31.97%)
0a6c11c — 2025-12-04 00:39:100.07 (n/a)0.06 (n/a)0.06 (n/a)0.04 (n/a)0.01 (n/a)139.80 (n/a)100.22 (n/a)82.20 (n/a)76.20 (n/a)26.31 (n/a)

dequant_1_cols_2_channels_2048_tile_1024_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.07 (+24.45%)0.05 (+9.74%)0.05 (-2.94%)0.03 (+16.88%)0.01 (+14.24%)163.70 (-14.47%)114.66 (-9.57%)111.30 (+3.06%)76.50 (-19.64%)31.63 (-21.04%)
0f0da0b — 2026-01-14 16:31:000.06 (n/a)0.04 (n/a)0.05 (n/a)0.03 (n/a)0.01 (n/a)191.40 (n/a)126.80 (n/a)108.00 (n/a)95.20 (n/a)40.06 (n/a)

dequant_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.07 (+5.66%)0.05 (+1.21%)0.05 (+0.61%)0.04 (-15.02%)0.01 (+58.42%)143.40 (+17.64%)101.02 (+1.49%)97.60 (-0.61%)78.90 (-5.40%)22.56 (+78.52%)
0a6c11c — 2025-12-04 00:39:100.06 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)121.90 (n/a)99.54 (n/a)98.20 (n/a)83.40 (n/a)12.64 (n/a)

dequant_2_cols_1_channels_2048_tile_1024_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.05 (+2.33%)0.05 (+22.41%)0.05 (+32.19%)0.04 (+7.46%)0.01 (-3.28%)147.80 (-6.93%)113.14 (-18.51%)106.50 (-24.31%)101.40 (-2.31%)19.61 (-8.79%)
0f0da0b — 2026-01-14 16:31:000.05 (n/a)0.04 (n/a)0.04 (n/a)0.03 (n/a)0.01 (n/a)158.80 (n/a)138.84 (n/a)140.70 (n/a)103.80 (n/a)21.51 (n/a)

dequant_2_cols_2_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.05 (+17.06%)0.04 (+8.98%)0.05 (+21.37%)0.03 (-11.65%)0.01 (+301.89%)159.70 (+13.18%)126.32 (-5.68%)109.00 (-17.67%)106.80 (-14.56%)22.48 (+280.15%)
0a6c11c — 2025-12-04 00:39:100.04 (n/a)0.04 (n/a)0.04 (n/a)0.04 (n/a)0.00 (n/a)141.10 (n/a)133.92 (n/a)132.40 (n/a)125.00 (n/a)5.91 (n/a)

dequant_2_cols_2_channels_2048_tile_512_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.05 (+14.83%)0.04 (+10.04%)0.05 (+21.96%)0.03 (-12.01%)0.01 (+143.96%)158.30 (+13.64%)121.38 (-6.49%)106.20 (-17.99%)98.20 (-12.94%)26.13 (+141.52%)
0f0da0b — 2026-01-14 16:31:000.05 (n/a)0.04 (n/a)0.04 (n/a)0.04 (n/a)0.00 (n/a)139.30 (n/a)129.80 (n/a)129.50 (n/a)112.80 (n/a)10.82 (n/a)

dequant_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.06 (+20.24%)0.05 (-3.19%)0.05 (-6.73%)0.03 (-20.69%)0.01 (+194.26%)156.10 (+26.09%)120.86 (+7.15%)114.90 (+7.18%)88.30 (-16.85%)24.40 (+214.82%)
0a6c11c — 2025-12-04 00:39:100.05 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.00 (n/a)123.80 (n/a)112.80 (n/a)107.20 (n/a)106.20 (n/a)7.75 (n/a)

dequant_4_cols_1_channels_2048_tile_512_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.06 (+28.85%)0.05 (+16.79%)0.05 (+23.60%)0.03 (-12.66%)0.01 (+157.83%)164.80 (+14.52%)113.62 (-10.20%)103.20 (-19.12%)84.80 (-22.42%)32.55 (+130.11%)
0f0da0b — 2026-01-14 16:31:000.05 (n/a)0.04 (n/a)0.04 (n/a)0.04 (n/a)0.00 (n/a)143.90 (n/a)126.52 (n/a)127.60 (n/a)109.30 (n/a)14.15 (n/a)

dequant_4_cols_2_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.05 (-6.09%)0.04 (-8.04%)0.04 (-12.46%)0.03 (-0.41%)0.01 (-20.19%)168.10 (+0.42%)135.74 (+7.44%)132.20 (+14.26%)107.40 (+6.44%)22.91 (-12.67%)
0a6c11c — 2025-12-04 00:39:100.05 (n/a)0.04 (n/a)0.05 (n/a)0.03 (n/a)0.01 (n/a)167.40 (n/a)126.34 (n/a)115.70 (n/a)100.90 (n/a)26.23 (n/a)

dequant_4_cols_2_channels_2048_tile_256_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.07 (+42.90%)0.05 (+60.94%)0.06 (+84.55%)0.03 (+50.15%)0.01 (+27.03%)188.40 (-33.40%)112.72 (-39.08%)93.80 (-45.84%)79.70 (-29.96%)43.58 (-36.67%)
0f0da0b — 2026-01-14 16:31:000.05 (n/a)0.03 (n/a)0.03 (n/a)0.02 (n/a)0.01 (n/a)282.90 (n/a)185.04 (n/a)173.20 (n/a)113.80 (n/a)68.82 (n/a)

dequant_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.04 (-13.65%)0.04 (-13.67%)0.04 (-18.56%)0.03 (-13.52%)0.00 (-23.88%)170.40 (+15.68%)140.12 (+15.33%)135.80 (+22.78%)119.60 (+15.78%)17.94 (+1.87%)
0a6c11c — 2025-12-04 00:39:100.05 (n/a)0.04 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)147.30 (n/a)121.50 (n/a)110.60 (n/a)103.30 (n/a)17.61 (n/a)

dequant_8_cols_1_channels_2048_tile_256_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.04 (-5.92%)0.04 (-11.07%)0.04 (-12.74%)0.03 (-9.17%)0.00 (-4.85%)166.90 (+10.09%)138.66 (+12.51%)135.90 (+14.59%)117.60 (+6.33%)17.94 (+10.10%)
0f0da0b — 2026-01-14 16:31:000.05 (n/a)0.04 (n/a)0.04 (n/a)0.03 (n/a)0.00 (n/a)151.60 (n/a)123.24 (n/a)118.60 (n/a)110.60 (n/a)16.30 (n/a)

dequant_8_cols_2_channels_2048_tile_128

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.05 (+17.21%)0.03 (-1.70%)0.03 (-3.04%)0.03 (-16.20%)0.01 (+95.59%)198.20 (+19.33%)160.00 (+4.92%)166.20 (+3.17%)106.50 (-14.66%)29.86 (+87.19%)
0a6c11c — 2025-12-04 00:39:100.04 (n/a)0.03 (n/a)0.03 (n/a)0.03 (n/a)0.00 (n/a)166.10 (n/a)152.50 (n/a)161.10 (n/a)124.80 (n/a)15.95 (n/a)

dequant_8_cols_2_channels_2048_tile_128_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.05 (+10.58%)0.04 (+11.15%)0.04 (+32.75%)0.02 (-15.30%)0.01 (+23.92%)229.90 (+18.08%)145.60 (-7.34%)131.00 (-24.67%)106.20 (-9.54%)49.03 (+42.68%)
0f0da0b — 2026-01-14 16:31:000.04 (n/a)0.03 (n/a)0.03 (n/a)0.03 (n/a)0.01 (n/a)194.70 (n/a)157.14 (n/a)173.90 (n/a)117.40 (n/a)34.37 (n/a)

eltwise_add_1_cols_2_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.17 (+59.76%)0.12 (+36.17%)0.10 (+5.69%)0.08 (+17.86%)0.04 (+129.37%)155.80 (-15.19%)109.78 (-22.78%)120.50 (-5.34%)73.80 (-37.40%)34.38 (+16.64%)
0f0da0b — 2026-01-14 16:31:000.10 (n/a)0.09 (n/a)0.10 (n/a)0.07 (n/a)0.02 (n/a)183.70 (n/a)142.16 (n/a)127.30 (n/a)117.90 (n/a)29.48 (n/a)

eltwise_add_2_cols_2_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.15 (+24.10%)0.13 (+23.91%)0.13 (+30.32%)0.08 (+16.80%)0.03 (+31.09%)146.70 (-14.36%)101.20 (-18.68%)91.70 (-23.26%)81.40 (-19.41%)26.00 (-6.71%)
0f0da0b — 2026-01-14 16:31:000.12 (n/a)0.10 (n/a)0.10 (n/a)0.07 (n/a)0.02 (n/a)171.30 (n/a)124.44 (n/a)119.50 (n/a)101.00 (n/a)27.87 (n/a)

eltwise_add_4_cols_2_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.17 (+52.81%)0.15 (+77.83%)0.15 (+81.17%)0.11 (+106.70%)0.02 (+14.06%)116.00 (-51.65%)86.46 (-45.77%)80.90 (-44.82%)72.70 (-34.56%)17.22 (-64.39%)
0f0da0b — 2026-01-14 16:31:000.11 (n/a)0.08 (n/a)0.08 (n/a)0.05 (n/a)0.02 (n/a)239.90 (n/a)159.44 (n/a)146.60 (n/a)111.10 (n/a)48.37 (n/a)

eltwise_add_8_cols_2_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.16 (+58.53%)0.14 (+73.02%)0.14 (+70.95%)0.11 (+188.94%)0.02 (-23.24%)109.50 (-65.38%)90.00 (-48.14%)86.00 (-41.50%)75.30 (-36.93%)13.44 (-83.58%)
0f0da0b — 2026-01-14 16:31:000.10 (n/a)0.08 (n/a)0.08 (n/a)0.04 (n/a)0.03 (n/a)316.30 (n/a)173.56 (n/a)147.00 (n/a)119.40 (n/a)81.81 (n/a)

eltwise_mul_1_cols_2_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.14 (+42.19%)0.13 (+64.04%)0.13 (+74.51%)0.10 (+75.35%)0.01 (+0.96%)118.10 (-42.97%)97.88 (-39.87%)92.60 (-42.70%)90.40 (-29.70%)11.55 (-59.47%)
0f0da0b — 2026-01-14 16:31:000.10 (n/a)0.08 (n/a)0.08 (n/a)0.06 (n/a)0.01 (n/a)207.10 (n/a)162.78 (n/a)161.60 (n/a)128.60 (n/a)28.50 (n/a)

eltwise_mul_2_cols_2_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.13 (+44.68%)0.12 (+43.52%)0.13 (+47.71%)0.09 (+35.56%)0.02 (+80.76%)130.00 (-26.26%)102.10 (-29.85%)93.30 (-32.34%)91.10 (-30.88%)16.53 (-8.94%)
0f0da0b — 2026-01-14 16:31:000.09 (n/a)0.09 (n/a)0.09 (n/a)0.07 (n/a)0.01 (n/a)176.30 (n/a)145.54 (n/a)137.90 (n/a)131.80 (n/a)18.15 (n/a)

eltwise_mul_4_cols_2_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.17 (+63.62%)0.15 (+73.87%)0.15 (+72.82%)0.11 (+104.66%)0.02 (+10.66%)114.00 (-51.16%)86.18 (-44.46%)81.50 (-42.12%)74.20 (-38.88%)15.85 (-65.75%)
0f0da0b — 2026-01-14 16:31:000.10 (n/a)0.08 (n/a)0.09 (n/a)0.05 (n/a)0.02 (n/a)233.40 (n/a)155.18 (n/a)140.80 (n/a)121.40 (n/a)46.29 (n/a)

eltwise_mul_8_cols_2_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.13 (+46.12%)0.11 (+59.06%)0.12 (+66.26%)0.08 (+48.55%)0.02 (+55.56%)154.80 (-32.70%)112.28 (-36.89%)104.60 (-39.85%)95.50 (-31.59%)24.14 (-26.88%)
0f0da0b — 2026-01-14 16:31:000.09 (n/a)0.07 (n/a)0.07 (n/a)0.05 (n/a)0.01 (n/a)230.00 (n/a)177.92 (n/a)173.90 (n/a)139.60 (n/a)33.01 (n/a)

embedding_dim_2048-hidden_dim_2048

No metrics available.

gelu_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.11 (+53.19%)0.10 (+56.21%)0.11 (+82.01%)0.08 (+32.44%)0.02 (+220.41%)101.60 (-24.52%)83.80 (-34.61%)72.40 (-45.11%)72.20 (-34.72%)15.75 (+58.78%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.06 (n/a)0.06 (n/a)0.06 (n/a)0.01 (n/a)134.60 (n/a)128.16 (n/a)131.90 (n/a)110.60 (n/a)9.92 (n/a)

gelu_1_cols_2_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.12 (+63.09%)0.11 (+76.43%)0.12 (+100.14%)0.07 (+62.57%)0.02 (+62.52%)110.00 (-38.48%)80.26 (-43.34%)69.10 (-50.00%)66.40 (-38.69%)18.54 (-39.34%)
0f0da0b — 2026-01-14 16:31:000.08 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)178.80 (n/a)141.66 (n/a)138.20 (n/a)108.30 (n/a)30.55 (n/a)

gelu_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.12 (+51.06%)0.10 (+80.20%)0.12 (+120.49%)0.06 (+32.94%)0.03 (+101.39%)137.80 (-24.74%)85.02 (-42.29%)68.20 (-54.65%)67.80 (-33.79%)30.24 (+4.82%)
0f0da0b — 2026-01-14 16:31:000.08 (n/a)0.06 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)183.10 (n/a)147.32 (n/a)150.40 (n/a)102.40 (n/a)28.85 (n/a)

gelu_2_cols_2_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.12 (+49.86%)0.11 (+81.45%)0.12 (+78.91%)0.08 (+119.08%)0.02 (+8.63%)107.80 (-54.38%)76.80 (-47.80%)69.50 (-44.13%)67.10 (-33.23%)17.40 (-67.51%)
0f0da0b — 2026-01-14 16:31:000.08 (n/a)0.06 (n/a)0.07 (n/a)0.03 (n/a)0.02 (n/a)236.30 (n/a)147.12 (n/a)124.40 (n/a)100.50 (n/a)53.56 (n/a)

gelu_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.13 (+73.67%)0.11 (+92.11%)0.12 (+121.10%)0.05 (+47.52%)0.03 (+98.64%)153.10 (-32.20%)85.56 (-45.77%)69.30 (-54.77%)63.00 (-42.47%)37.94 (-17.18%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.06 (n/a)0.05 (n/a)0.04 (n/a)0.02 (n/a)225.80 (n/a)157.78 (n/a)153.20 (n/a)109.50 (n/a)45.81 (n/a)

gelu_4_cols_2_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.12 (+87.02%)0.11 (+117.65%)0.12 (+144.58%)0.07 (+103.45%)0.02 (+63.92%)110.50 (-50.82%)79.80 (-54.61%)70.50 (-59.08%)70.00 (-46.56%)17.48 (-56.79%)
0f0da0b — 2026-01-14 16:31:000.06 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)224.70 (n/a)175.80 (n/a)172.30 (n/a)131.00 (n/a)40.46 (n/a)

gelu_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.12 (+61.36%)0.10 (+65.53%)0.11 (+79.09%)0.04 (-5.63%)0.03 (+170.92%)208.20 (+5.95%)98.74 (-30.26%)71.90 (-44.13%)70.20 (-38.04%)61.20 (+79.92%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.06 (n/a)0.06 (n/a)0.04 (n/a)0.01 (n/a)196.50 (n/a)141.58 (n/a)128.70 (n/a)113.30 (n/a)34.01 (n/a)

gelu_8_cols_2_channels_2048_tile_128

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.10 (+64.20%)0.09 (+72.61%)0.10 (+77.79%)0.06 (+39.91%)0.02 (+119.55%)133.40 (-28.55%)93.44 (-40.99%)82.30 (-43.75%)81.90 (-39.11%)22.47 (-5.49%)
0f0da0b — 2026-01-14 16:31:000.06 (n/a)0.05 (n/a)0.06 (n/a)0.04 (n/a)0.01 (n/a)186.70 (n/a)158.34 (n/a)146.30 (n/a)134.50 (n/a)23.77 (n/a)

gemm_1792x896x1152_64x32x48_8cols_ccolmaj

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
1531781 — 2026-01-23 18:49:245.15 (+4.03%)4.90 (+9.86%)4.99 (+12.67%)4.27 (+4.36%)0.36 (+11.38%)2201.70 (-4.17%)1929.94 (-8.91%)1885.00 (-11.25%)1826.70 (-3.87%)154.05 (+4.53%)2025.19 (+4.03%)1925.76 (+9.86%)1962.51 (+12.67%)1680.20 (+4.36%)139.97 (+11.38%)
0f0da0b — 2026-01-14 16:31:004.95 (n/a)4.46 (n/a)4.43 (n/a)4.09 (n/a)0.32 (n/a)2297.60 (n/a)2118.80 (n/a)2123.90 (n/a)1900.20 (n/a)147.37 (n/a)1946.79 (n/a)1752.94 (n/a)1741.75 (n/a)1610.08 (n/a)125.67 (n/a)

gemm_192x384x64_48x96x16_4cols

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
1531781 — 2026-01-23 18:49:242.36 (+52.61%)1.85 (+39.10%)1.78 (+41.59%)1.32 (+14.31%)0.39 (+132.56%)167.80 (-12.51%)124.24 (-26.22%)124.40 (-29.36%)93.70 (-34.52%)28.11 (+35.82%)100.67 (+52.61%)78.94 (+39.10%)75.88 (+41.59%)56.25 (+14.31%)16.73 (+132.56%)
0f0da0b — 2026-01-14 16:31:001.55 (n/a)1.33 (n/a)1.26 (n/a)1.15 (n/a)0.17 (n/a)191.80 (n/a)168.40 (n/a)176.10 (n/a)143.10 (n/a)20.70 (n/a)65.96 (n/a)56.75 (n/a)53.59 (n/a)49.20 (n/a)7.20 (n/a)

gemm_192x384x64_48x96x16_4cols_bcolmaj_ccolmaj

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
1531781 — 2026-01-23 18:49:242.22 (+30.89%)2.06 (+54.21%)2.20 (+75.49%)1.60 (+35.74%)0.26 (+25.57%)138.20 (-26.33%)109.20 (-35.27%)100.50 (-43.03%)99.80 (-23.58%)16.54 (-28.10%)94.54 (+30.89%)87.77 (+54.21%)93.91 (+75.49%)68.30 (+35.74%)11.26 (+25.57%)
0f0da0b — 2026-01-14 16:31:001.69 (n/a)1.33 (n/a)1.25 (n/a)1.18 (n/a)0.21 (n/a)187.60 (n/a)168.70 (n/a)176.40 (n/a)130.60 (n/a)23.00 (n/a)72.23 (n/a)56.92 (n/a)53.51 (n/a)50.31 (n/a)8.97 (n/a)

gemm_2048x2048x2048_64x64x32_8_cols_0_bcolmaj_0_ccolmaj_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)GFLOP/s (max)GFLOP/s (mean)GFLOP/s (median)GFLOP/s (min)GFLOP/s (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3a053d3 — 2025-11-26 01:12:163.72 (-0.11%)3.55 (-2.75%)3.56 (-2.75%)3.36 (-5.33%)0.12 (+76.10%)2725.23 (-0.11%)2602.52 (-2.75%)2612.11 (-2.75%)2459.89 (-5.33%)86.27 (+76.10%)6984.00 (+5.63%)6608.60 (+2.91%)6577.00 (+2.83%)6304.00 (+0.11%)221.87 (+87.15%)
9e98049 — 2025-11-25 20:50:553.72 (n/a)3.65 (n/a)3.66 (n/a)3.54 (n/a)0.07 (n/a)2728.26 (n/a)2676.06 (n/a)2686.03 (n/a)2598.29 (n/a)48.99 (n/a)6612.00 (n/a)6422.00 (n/a)6396.00 (n/a)6297.00 (n/a)118.55 (n/a)

gemm_2048x2048x2048_64x64x32_8_cols_0_bcolmaj_1_ccolmaj_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)GFLOP/s (max)GFLOP/s (mean)GFLOP/s (median)GFLOP/s (min)GFLOP/s (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3a053d3 — 2025-11-26 01:12:163.40 (-3.14%)3.38 (-0.82%)3.38 (-0.82%)3.35 (+4.93%)0.02 (-84.51%)2493.81 (-3.14%)2475.27 (-0.82%)2479.06 (-0.82%)2453.92 (+4.93%)13.33 (-84.51%)7001.00 (-4.70%)6940.80 (+0.71%)6930.00 (+0.83%)6889.00 (+3.24%)37.43 (-84.76%)
9e98049 — 2025-11-25 20:50:553.51 (n/a)3.40 (n/a)3.41 (n/a)3.19 (n/a)0.12 (n/a)2574.53 (n/a)2495.72 (n/a)2499.62 (n/a)2338.67 (n/a)86.02 (n/a)7346.00 (n/a)6892.20 (n/a)6873.00 (n/a)6673.00 (n/a)245.57 (n/a)

gemm_2048x2048x2048_64x64x32_8_cols_1_bcolmaj_0_ccolmaj_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)GFLOP/s (max)GFLOP/s (mean)GFLOP/s (median)GFLOP/s (min)GFLOP/s (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3a053d3 — 2025-11-26 01:12:163.02 (+0.52%)2.87 (-2.05%)2.86 (-2.24%)2.70 (-6.16%)0.11 (+131.66%)2216.19 (+0.52%)2104.44 (-2.05%)2094.08 (-2.24%)1980.84 (-6.16%)80.25 (+131.67%)8673.00 (+6.56%)8175.60 (+2.21%)8204.00 (+2.29%)7752.00 (-0.51%)314.03 (+145.36%)
9e98049 — 2025-11-25 20:50:553.01 (n/a)2.93 (n/a)2.92 (n/a)2.88 (n/a)0.05 (n/a)2204.81 (n/a)2148.41 (n/a)2142.13 (n/a)2110.81 (n/a)34.64 (n/a)8139.00 (n/a)7998.60 (n/a)8020.00 (n/a)7792.00 (n/a)127.99 (n/a)

gemm_2048x2048x2048_64x64x64_1cols

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
1531781 — 2026-01-23 18:49:240.52 (-0.75%)0.52 (-0.07%)0.52 (+0.01%)0.52 (+0.22%)0.00 (-86.34%)48415.10 (-0.22%)48370.52 (+0.07%)48362.20 (-0.01%)48344.40 (+0.76%)29.40 (-86.25%)355.36 (-0.75%)355.17 (-0.07%)355.23 (+0.01%)354.85 (+0.22%)0.22 (-86.34%)
0f0da0b — 2026-01-14 16:31:000.52 (n/a)0.52 (n/a)0.52 (n/a)0.52 (n/a)0.00 (n/a)48520.80 (n/a)48335.84 (n/a)48365.40 (n/a)47981.00 (n/a)213.79 (n/a)358.06 (n/a)355.43 (n/a)355.21 (n/a)354.07 (n/a)1.58 (n/a)

gemm_2048x2048x2048_64x64x64_2_cols_0_bcolmaj_0_ccolmaj_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:491.04 (+0.94%)1.03 (+0.23%)1.03 (+0.09%)1.02 (+0.07%)0.00 (+197.51%)24576.00 (-0.07%)24485.82 (-0.23%)24530.40 (-0.09%)24266.40 (-0.94%)111.26 (+194.35%)
0a6c11c — 2025-12-04 00:39:101.03 (n/a)1.03 (n/a)1.03 (n/a)1.02 (n/a)0.00 (n/a)24594.30 (n/a)24542.14 (n/a)24551.90 (n/a)24495.50 (n/a)37.80 (n/a)

gemm_2048x2048x2048_64x64x64_2_cols_0_bcolmaj_0_ccolmaj_0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
a4b6ffe — 2026-01-05 19:24:051.04 (+0.28%)1.03 (-0.15%)1.03 (+0.05%)1.02 (-0.52%)0.01 (+42.72%)24719.80 (+0.52%)24512.42 (+0.16%)24529.20 (-0.05%)24092.60 (-0.28%)252.92 (+43.15%)713.08 (+0.28%)700.92 (-0.15%)700.38 (+0.05%)694.98 (-0.52%)7.30 (+42.71%)
50ad37c — 2025-12-29 20:35:531.04 (n/a)1.03 (n/a)1.03 (n/a)1.02 (n/a)0.01 (n/a)24592.30 (n/a)24473.70 (n/a)24540.90 (n/a)24160.80 (n/a)176.68 (n/a)711.06 (n/a)702.00 (n/a)700.05 (n/a)698.59 (n/a)5.12 (n/a)

gemm_2048x2048x2048_64x64x64_2_cols_0_bcolmaj_1_ccolmaj_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.99 (+1.69%)0.97 (+0.29%)0.97 (-0.04%)0.97 (-0.15%)0.01 (+434.92%)26025.30 (+0.15%)25839.34 (-0.29%)25916.70 (+0.04%)25446.00 (-1.67%)202.05 (+425.44%)
0a6c11c — 2025-12-04 00:39:100.97 (n/a)0.97 (n/a)0.97 (n/a)0.97 (n/a)0.00 (n/a)25986.90 (n/a)25913.52 (n/a)25905.70 (n/a)25877.00 (n/a)38.45 (n/a)

gemm_2048x2048x2048_64x64x64_2_cols_0_bcolmaj_1_ccolmaj_0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
a4b6ffe — 2026-01-05 19:24:050.97 (-0.01%)0.97 (+0.03%)0.97 (+0.04%)0.97 (+0.00%)0.00 (-6.43%)25927.90 (-0.00%)25903.20 (-0.03%)25898.60 (-0.04%)25892.80 (+0.01%)14.05 (-6.56%)663.50 (-0.01%)663.23 (+0.03%)663.35 (+0.04%)662.60 (+0.00%)0.36 (-6.43%)
50ad37c — 2025-12-29 20:35:530.97 (n/a)0.97 (n/a)0.97 (n/a)0.97 (n/a)0.00 (n/a)25928.10 (n/a)25910.14 (n/a)25908.10 (n/a)25891.20 (n/a)15.04 (n/a)663.54 (n/a)663.06 (n/a)663.11 (n/a)662.60 (n/a)0.38 (n/a)

gemm_2048x2048x2048_64x64x64_2_cols_1_bcolmaj_0_ccolmaj_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.92 (-1.02%)0.92 (-0.66%)0.92 (-0.10%)0.91 (-0.78%)0.01 (-4.54%)27751.40 (+0.79%)27480.74 (+0.66%)27354.80 (+0.10%)27320.50 (+1.03%)182.69 (-2.84%)
0a6c11c — 2025-12-04 00:39:100.93 (n/a)0.92 (n/a)0.92 (n/a)0.91 (n/a)0.01 (n/a)27534.80 (n/a)27299.32 (n/a)27326.70 (n/a)27042.30 (n/a)188.04 (n/a)

gemm_2048x2048x2048_64x64x64_2_cols_1_bcolmaj_0_ccolmaj_0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
a4b6ffe — 2026-01-05 19:24:050.92 (+0.11%)0.91 (-0.12%)0.91 (-0.21%)0.91 (-0.24%)0.00 (+30.34%)27723.50 (+0.24%)27548.94 (+0.12%)27522.90 (+0.21%)27403.60 (-0.11%)124.26 (+30.58%)626.92 (+0.11%)623.62 (-0.12%)624.20 (-0.21%)619.69 (-0.24%)2.81 (+30.34%)
50ad37c — 2025-12-29 20:35:530.92 (n/a)0.91 (n/a)0.92 (n/a)0.91 (n/a)0.00 (n/a)27657.10 (n/a)27515.58 (n/a)27464.70 (n/a)27434.60 (n/a)95.16 (n/a)626.21 (n/a)624.37 (n/a)625.52 (n/a)621.17 (n/a)2.16 (n/a)

gemm_2048x2048x2048_64x64x64_2cols_bcolmaj

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
1531781 — 2026-01-23 18:49:240.93 (+1.26%)0.91 (+0.29%)0.91 (+0.24%)0.90 (-0.34%)0.01 (+132.37%)27826.70 (+0.34%)27533.84 (-0.28%)27519.90 (-0.24%)27105.20 (-1.24%)272.35 (+129.91%)633.82 (+1.26%)624.00 (+0.29%)624.27 (+0.24%)617.39 (-0.34%)6.21 (+132.36%)
0f0da0b — 2026-01-14 16:31:000.92 (n/a)0.91 (n/a)0.91 (n/a)0.91 (n/a)0.00 (n/a)27731.40 (n/a)27612.36 (n/a)27585.70 (n/a)27446.50 (n/a)118.46 (n/a)625.94 (n/a)622.19 (n/a)622.78 (n/a)619.51 (n/a)2.67 (n/a)

gemm_2048x2048x2048_64x64x64_8_cols_0_bcolmaj_0_ccolmaj_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)GFLOP/s (max)GFLOP/s (mean)GFLOP/s (median)GFLOP/s (min)GFLOP/s (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
31ab24e — 2025-11-22 02:01:563.65 (-1.06%)3.59 (+0.83%)3.63 (-0.26%)3.45 (+2.66%)0.07 (-44.71%)2678.08 (-1.06%)2632.13 (+0.83%)2658.60 (-0.26%)2528.31 (+2.66%)53.86 (-44.71%)6795.00 (-2.59%)6529.80 (-0.92%)6462.00 (+0.26%)6415.00 (+1.07%)137.14 (-45.27%)
d9a64c5 — 2025-11-21 20:49:353.69 (n/a)3.56 (n/a)3.64 (n/a)3.36 (n/a)0.13 (n/a)2706.77 (n/a)2610.50 (n/a)2665.61 (n/a)2462.71 (n/a)97.41 (n/a)6976.00 (n/a)6590.40 (n/a)6445.00 (n/a)6347.00 (n/a)250.58 (n/a)

gemm_2048x2048x2048_64x64x64_8_cols_0_bcolmaj_0_ccolmaj_0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
d48746f — 2026-01-12 22:17:103.98 (+0.20%)3.84 (-2.95%)3.76 (-5.27%)3.72 (-4.97%)0.13 (+473.97%)6760.00 (+5.23%)6559.50 (+3.13%)6697.10 (+5.56%)6321.00 (-0.20%)218.68 (+499.40%)2717.91 (+0.20%)2621.44 (-2.95%)2565.27 (-5.27%)2541.40 (-4.97%)88.37 (+473.97%)
331dcca — 2026-01-08 18:34:103.97 (n/a)3.96 (n/a)3.97 (n/a)3.92 (n/a)0.02 (n/a)6423.80 (n/a)6360.32 (n/a)6344.30 (n/a)6333.90 (n/a)36.48 (n/a)2712.39 (n/a)2701.17 (n/a)2707.91 (n/a)2674.40 (n/a)15.40 (n/a)

gemm_2048x2048x2048_64x64x64_8_cols_0_bcolmaj_1_ccolmaj_0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
d48746f — 2026-01-12 22:17:103.75 (+0.28%)3.70 (+1.25%)3.71 (-0.21%)3.65 (+4.77%)0.04 (-66.90%)6901.80 (-4.56%)6795.46 (-1.30%)6790.20 (+0.21%)6715.90 (-0.28%)67.05 (-68.40%)2558.08 (+0.28%)2528.33 (+1.25%)2530.10 (-0.21%)2489.18 (+4.77%)24.81 (-66.90%)
331dcca — 2026-01-08 18:34:103.74 (n/a)3.66 (n/a)3.71 (n/a)3.48 (n/a)0.11 (n/a)7231.20 (n/a)6885.18 (n/a)6775.70 (n/a)6734.90 (n/a)212.21 (n/a)2550.87 (n/a)2497.04 (n/a)2535.50 (n/a)2375.78 (n/a)74.96 (n/a)

gemm_2048x2048x2048_64x64x64_8_cols_1_bcolmaj_0_ccolmaj_0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
d48746f — 2026-01-12 22:17:103.62 (-0.77%)3.54 (+0.14%)3.60 (-0.04%)3.44 (+2.17%)0.09 (-25.77%)7307.40 (-2.13%)7103.46 (-0.18%)6982.50 (+0.04%)6948.20 (+0.77%)182.95 (-26.55%)2472.55 (-0.77%)2419.80 (+0.14%)2460.41 (-0.04%)2351.04 (+2.17%)61.76 (-25.77%)
331dcca — 2026-01-08 18:34:103.65 (n/a)3.54 (n/a)3.61 (n/a)3.37 (n/a)0.12 (n/a)7466.20 (n/a)7116.40 (n/a)6979.60 (n/a)6895.10 (n/a)249.09 (n/a)2491.61 (n/a)2416.45 (n/a)2461.45 (n/a)2301.00 (n/a)83.20 (n/a)

gemm_2048x2048x2048_64x64x64_8cols_bcolmaj_ccolmaj

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
1531781 — 2026-01-23 18:49:243.46 (+1.18%)3.42 (+2.60%)3.45 (+2.90%)3.29 (+2.92%)0.07 (-25.59%)7651.20 (-2.83%)7368.34 (-2.57%)7300.80 (-2.82%)7263.70 (-1.17%)162.53 (-28.22%)2365.17 (+1.18%)2332.46 (+2.60%)2353.14 (+2.90%)2245.39 (+2.92%)50.15 (-25.59%)
0f0da0b — 2026-01-14 16:31:003.42 (n/a)3.33 (n/a)3.35 (n/a)3.20 (n/a)0.10 (n/a)7874.40 (n/a)7562.58 (n/a)7512.80 (n/a)7349.50 (n/a)226.42 (n/a)2337.56 (n/a)2273.31 (n/a)2286.73 (n/a)2181.74 (n/a)67.40 (n/a)

gemm_384x1536x1792_32x48x64_4cols_bcolmaj

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
1531781 — 2026-01-23 18:49:244.35 (-0.80%)4.24 (+3.81%)4.33 (+9.58%)3.85 (-0.98%)0.22 (-4.90%)2094.90 (+0.99%)1905.64 (-3.69%)1862.00 (-8.75%)1852.00 (+0.81%)105.88 (-3.06%)1141.45 (-0.80%)1111.86 (+3.81%)1135.29 (+9.58%)1009.08 (-0.98%)57.51 (-4.90%)
0f0da0b — 2026-01-14 16:31:004.39 (n/a)4.08 (n/a)3.95 (n/a)3.89 (n/a)0.23 (n/a)2074.30 (n/a)1978.70 (n/a)2040.50 (n/a)1837.10 (n/a)109.22 (n/a)1150.67 (n/a)1071.02 (n/a)1036.00 (n/a)1019.11 (n/a)60.48 (n/a)

gemm_896x1792x640_32x64x80_8cols_ccolmaj

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
1531781 — 2026-01-23 18:49:246.80 (+27.89%)6.43 (+24.70%)6.70 (+30.77%)5.33 (+5.74%)0.62 (+474.69%)1247.10 (-5.44%)1042.70 (-19.16%)992.90 (-23.53%)978.30 (-21.81%)114.59 (+329.89%)2100.72 (+27.89%)1987.76 (+24.70%)2069.85 (+30.77%)1647.93 (+5.74%)190.83 (+474.69%)
0f0da0b — 2026-01-14 16:31:005.32 (n/a)5.16 (n/a)5.12 (n/a)5.04 (n/a)0.11 (n/a)1318.80 (n/a)1289.76 (n/a)1298.50 (n/a)1251.20 (n/a)26.66 (n/a)1642.57 (n/a)1594.05 (n/a)1582.78 (n/a)1558.44 (n/a)33.21 (n/a)

input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
82cf4ab — 2026-02-09 20:43:010.08 (n/a)0.05 (n/a)0.05 (n/a)0.03 (n/a)0.01 (n/a)277.80 (n/a)162.77 (n/a)155.10 (n/a)105.90 (n/a)42.66 (n/a)

input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048-group_size_32

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
82cf4ab — 2026-02-09 20:43:010.04 (n/a)0.03 (n/a)0.03 (n/a)0.02 (n/a)0.01 (n/a)214.80 (n/a)176.28 (n/a)178.60 (n/a)125.60 (n/a)32.25 (n/a)

input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048-weighted_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
82cf4ab — 2026-02-09 20:43:010.05 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.00 (n/a)184.50 (n/a)164.50 (n/a)161.90 (n/a)152.10 (n/a)12.46 (n/a)

input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
82cf4ab — 2026-02-09 20:43:010.07 (n/a)0.05 (n/a)0.06 (n/a)0.04 (n/a)0.01 (n/a)210.00 (n/a)157.19 (n/a)144.95 (n/a)115.20 (n/a)30.43 (n/a)

input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024-group_size_32

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
82cf4ab — 2026-02-09 20:43:010.04 (n/a)0.03 (n/a)0.03 (n/a)0.03 (n/a)0.00 (n/a)176.20 (n/a)157.18 (n/a)152.60 (n/a)141.60 (n/a)16.47 (n/a)

input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024-weighted_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
82cf4ab — 2026-02-09 20:43:010.06 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)207.20 (n/a)168.68 (n/a)163.60 (n/a)146.90 (n/a)23.13 (n/a)

input_length_2048-num_aie_columns_1-num_channels_2-tile_size_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
82cf4ab — 2026-02-09 20:43:010.11 (n/a)0.08 (n/a)0.09 (n/a)0.06 (n/a)0.02 (n/a)206.00 (n/a)159.43 (n/a)144.50 (n/a)107.30 (n/a)35.66 (n/a)

input_length_2048-num_aie_columns_1-num_channels_2-tile_size_2048-scalar_factor_3.0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
82cf4ab — 2026-02-09 20:43:010.11 (n/a)0.09 (n/a)0.09 (n/a)0.06 (n/a)0.02 (n/a)207.20 (n/a)151.40 (n/a)138.80 (n/a)109.40 (n/a)37.94 (n/a)

input_length_2048-num_aie_columns_1-num_channels_2-tile_size_2048-weighted_True

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
82cf4ab — 2026-02-09 20:43:010.08 (n/a)0.07 (n/a)0.08 (n/a)0.05 (n/a)0.01 (n/a)237.90 (n/a)185.10 (n/a)162.10 (n/a)159.30 (n/a)35.05 (n/a)

input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
82cf4ab — 2026-02-09 20:43:010.09 (n/a)0.06 (n/a)0.06 (n/a)0.03 (n/a)0.01 (n/a)281.80 (n/a)155.12 (n/a)144.50 (n/a)91.30 (n/a)41.10 (n/a)

input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024-group_size_32

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
82cf4ab — 2026-02-09 20:43:010.05 (n/a)0.04 (n/a)0.04 (n/a)0.03 (n/a)0.01 (n/a)186.40 (n/a)137.42 (n/a)129.00 (n/a)115.40 (n/a)27.99 (n/a)

input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024-weighted_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
82cf4ab — 2026-02-09 20:43:010.05 (n/a)0.05 (n/a)0.05 (n/a)0.05 (n/a)0.00 (n/a)180.50 (n/a)164.18 (n/a)161.20 (n/a)155.40 (n/a)10.11 (n/a)

input_length_2048-num_aie_columns_2-num_channels_2-tile_size_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
82cf4ab — 2026-02-09 20:43:010.12 (n/a)0.08 (n/a)0.07 (n/a)0.06 (n/a)0.02 (n/a)213.90 (n/a)167.48 (n/a)168.55 (n/a)100.10 (n/a)39.84 (n/a)

input_length_2048-num_aie_columns_2-num_channels_2-tile_size_1024-scalar_factor_3.0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
82cf4ab — 2026-02-09 20:43:010.11 (n/a)0.09 (n/a)0.10 (n/a)0.06 (n/a)0.02 (n/a)200.30 (n/a)141.44 (n/a)119.90 (n/a)114.70 (n/a)36.53 (n/a)

input_length_2048-num_aie_columns_2-num_channels_2-tile_size_1024-weighted_True

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
82cf4ab — 2026-02-09 20:43:010.05 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.00 (n/a)236.00 (n/a)208.82 (n/a)202.70 (n/a)190.30 (n/a)17.62 (n/a)

input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
82cf4ab — 2026-02-09 20:43:010.09 (n/a)0.05 (n/a)0.04 (n/a)0.03 (n/a)0.02 (n/a)253.40 (n/a)183.86 (n/a)187.30 (n/a)92.50 (n/a)48.12 (n/a)

input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512-group_size_32

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
82cf4ab — 2026-02-09 20:43:010.03 (n/a)0.03 (n/a)0.03 (n/a)0.03 (n/a)0.00 (n/a)182.90 (n/a)168.68 (n/a)174.30 (n/a)151.90 (n/a)13.70 (n/a)

input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512-weighted_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
82cf4ab — 2026-02-09 20:43:010.07 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)206.90 (n/a)166.40 (n/a)162.60 (n/a)117.80 (n/a)33.10 (n/a)

input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
82cf4ab — 2026-02-09 20:43:010.07 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)223.60 (n/a)161.01 (n/a)152.60 (n/a)112.10 (n/a)33.06 (n/a)

input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512-group_size_32

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
82cf4ab — 2026-02-09 20:43:010.04 (n/a)0.03 (n/a)0.03 (n/a)0.03 (n/a)0.00 (n/a)200.70 (n/a)162.64 (n/a)151.40 (n/a)138.00 (n/a)26.07 (n/a)

input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512-weighted_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
82cf4ab — 2026-02-09 20:43:010.05 (n/a)0.04 (n/a)0.05 (n/a)0.03 (n/a)0.01 (n/a)299.50 (n/a)204.18 (n/a)180.20 (n/a)155.90 (n/a)59.35 (n/a)

input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
82cf4ab — 2026-02-09 20:43:010.06 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)226.50 (n/a)165.15 (n/a)165.95 (n/a)127.80 (n/a)30.66 (n/a)

input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256-group_size_32

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
82cf4ab — 2026-02-09 20:43:010.03 (n/a)0.03 (n/a)0.03 (n/a)0.02 (n/a)0.01 (n/a)225.30 (n/a)192.64 (n/a)198.10 (n/a)151.60 (n/a)33.77 (n/a)

input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256-weighted_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
82cf4ab — 2026-02-09 20:43:010.05 (n/a)0.04 (n/a)0.04 (n/a)0.04 (n/a)0.01 (n/a)231.70 (n/a)194.78 (n/a)189.80 (n/a)171.70 (n/a)25.45 (n/a)

input_length_2048-num_aie_columns_4-num_channels_2-tile_size_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
82cf4ab — 2026-02-09 20:43:010.11 (n/a)0.07 (n/a)0.07 (n/a)0.06 (n/a)0.02 (n/a)219.10 (n/a)175.70 (n/a)186.90 (n/a)107.50 (n/a)37.49 (n/a)

input_length_2048-num_aie_columns_4-num_channels_2-tile_size_512-scalar_factor_3.0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
82cf4ab — 2026-02-09 20:43:010.11 (n/a)0.08 (n/a)0.09 (n/a)0.06 (n/a)0.02 (n/a)194.80 (n/a)155.18 (n/a)143.80 (n/a)110.90 (n/a)37.14 (n/a)

input_length_2048-num_aie_columns_4-num_channels_2-tile_size_512-weighted_True

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
82cf4ab — 2026-02-09 20:43:010.06 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)219.60 (n/a)183.10 (n/a)187.40 (n/a)146.60 (n/a)26.43 (n/a)

input_length_2048-num_aie_columns_8-num_channels_1-tile_size_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
82cf4ab — 2026-02-09 20:43:010.07 (n/a)0.05 (n/a)0.05 (n/a)0.02 (n/a)0.01 (n/a)356.00 (n/a)187.23 (n/a)181.55 (n/a)118.70 (n/a)50.32 (n/a)

input_length_2048-num_aie_columns_8-num_channels_1-tile_size_256-group_size_32

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
82cf4ab — 2026-02-09 20:43:010.04 (n/a)0.03 (n/a)0.03 (n/a)0.02 (n/a)0.01 (n/a)237.20 (n/a)173.28 (n/a)155.00 (n/a)142.80 (n/a)39.75 (n/a)

input_length_2048-num_aie_columns_8-num_channels_1-tile_size_256-weighted_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
82cf4ab — 2026-02-09 20:43:010.05 (n/a)0.04 (n/a)0.05 (n/a)0.03 (n/a)0.01 (n/a)246.50 (n/a)191.00 (n/a)181.70 (n/a)157.80 (n/a)33.93 (n/a)

input_length_2048-num_aie_columns_8-num_channels_2-tile_size_128

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
82cf4ab — 2026-02-09 20:43:010.05 (n/a)0.04 (n/a)0.04 (n/a)0.02 (n/a)0.01 (n/a)420.10 (n/a)223.23 (n/a)206.70 (n/a)176.90 (n/a)72.41 (n/a)

input_length_2048-num_aie_columns_8-num_channels_2-tile_size_128-group_size_32

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
82cf4ab — 2026-02-09 20:43:010.03 (n/a)0.03 (n/a)0.03 (n/a)0.02 (n/a)0.00 (n/a)236.70 (n/a)192.52 (n/a)197.40 (n/a)162.80 (n/a)30.29 (n/a)

input_length_2048-num_aie_columns_8-num_channels_2-tile_size_128-weighted_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
82cf4ab — 2026-02-09 20:43:010.04 (n/a)0.03 (n/a)0.03 (n/a)0.02 (n/a)0.01 (n/a)409.80 (n/a)280.10 (n/a)239.40 (n/a)188.40 (n/a)96.46 (n/a)

input_length_2048-num_aie_columns_8-num_channels_2-tile_size_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
82cf4ab — 2026-02-09 20:43:010.10 (n/a)0.07 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)244.60 (n/a)187.51 (n/a)191.65 (n/a)124.00 (n/a)32.28 (n/a)

input_length_2048-num_aie_columns_8-num_channels_2-tile_size_256-scalar_factor_3.0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
82cf4ab — 2026-02-09 20:43:010.10 (n/a)0.08 (n/a)0.08 (n/a)0.07 (n/a)0.02 (n/a)186.00 (n/a)154.94 (n/a)160.70 (n/a)122.80 (n/a)28.69 (n/a)

input_length_2048-num_aie_columns_8-num_channels_2-tile_size_256-weighted_True

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
82cf4ab — 2026-02-09 20:43:010.04 (n/a)0.04 (n/a)0.04 (n/a)0.02 (n/a)0.01 (n/a)417.20 (n/a)249.82 (n/a)209.60 (n/a)197.90 (n/a)93.86 (n/a)

input_length_2048-num_cores_1-num_channels_1-bypass_False-tile_size_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
82cf4ab — 2026-02-09 20:43:010.07 (n/a)0.05 (n/a)0.04 (n/a)0.03 (n/a)0.01 (n/a)250.30 (n/a)176.58 (n/a)187.40 (n/a)123.30 (n/a)52.62 (n/a)

input_length_2048-num_cores_16-num_channels_2-bypass_False-tile_size_128

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
82cf4ab — 2026-02-09 20:43:010.05 (n/a)0.04 (n/a)0.04 (n/a)0.03 (n/a)0.01 (n/a)261.80 (n/a)212.50 (n/a)199.80 (n/a)171.70 (n/a)39.89 (n/a)

input_length_2048-num_cores_2-num_channels_1-bypass_False-tile_size_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
82cf4ab — 2026-02-09 20:43:010.07 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)223.20 (n/a)159.70 (n/a)152.10 (n/a)116.60 (n/a)41.93 (n/a)

input_length_2048-num_cores_2-num_channels_2-bypass_False-tile_size_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
82cf4ab — 2026-02-09 20:43:010.07 (n/a)0.06 (n/a)0.05 (n/a)0.05 (n/a)0.01 (n/a)166.70 (n/a)149.42 (n/a)151.40 (n/a)117.80 (n/a)19.09 (n/a)

input_length_2048-num_cores_4-num_channels_1-bypass_False-tile_size_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
82cf4ab — 2026-02-09 20:43:010.06 (n/a)0.05 (n/a)0.04 (n/a)0.04 (n/a)0.01 (n/a)228.60 (n/a)184.16 (n/a)188.40 (n/a)148.80 (n/a)32.65 (n/a)

input_length_2048-num_cores_4-num_channels_2-bypass_False-tile_size_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
82cf4ab — 2026-02-09 20:43:010.07 (n/a)0.06 (n/a)0.06 (n/a)0.04 (n/a)0.01 (n/a)208.60 (n/a)150.00 (n/a)140.70 (n/a)125.60 (n/a)33.72 (n/a)

input_length_2048-num_cores_8-num_channels_1-bypass_False-tile_size_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
82cf4ab — 2026-02-09 20:43:010.06 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)200.70 (n/a)177.92 (n/a)181.20 (n/a)146.90 (n/a)19.79 (n/a)

input_length_2048-num_cores_8-num_channels_2-bypass_False-tile_size_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
82cf4ab — 2026-02-09 20:43:010.06 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)224.10 (n/a)178.72 (n/a)177.90 (n/a)130.20 (n/a)36.47 (n/a)

input_length_32768-num_aie_columns_2-num_channels_2-tile_size_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
82cf4ab — 2026-02-09 20:43:011.19 (n/a)1.03 (n/a)1.04 (n/a)0.91 (n/a)0.11 (n/a)143.50 (n/a)128.58 (n/a)126.40 (n/a)110.10 (n/a)13.28 (n/a)

input_length_32768-num_aie_columns_2-num_channels_2-tile_size_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
82cf4ab — 2026-02-09 20:43:011.12 (n/a)0.86 (n/a)0.85 (n/a)0.62 (n/a)0.22 (n/a)212.70 (n/a)160.80 (n/a)154.50 (n/a)116.60 (n/a)41.43 (n/a)

input_length_32768-num_aie_columns_4-num_channels_4-tile_size_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
82cf4ab — 2026-02-09 20:43:011.12 (n/a)0.77 (n/a)0.73 (n/a)0.41 (n/a)0.33 (n/a)323.50 (n/a)201.22 (n/a)178.60 (n/a)116.80 (n/a)91.41 (n/a)

layer_norm_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.11 (+88.61%)0.09 (+61.18%)0.08 (+46.00%)0.07 (+61.81%)0.02 (+189.47%)119.70 (-38.20%)97.44 (-36.14%)100.50 (-31.54%)71.50 (-46.96%)22.67 (-5.36%)
0f0da0b — 2026-01-14 16:31:000.06 (n/a)0.05 (n/a)0.06 (n/a)0.04 (n/a)0.01 (n/a)193.70 (n/a)152.58 (n/a)146.80 (n/a)134.80 (n/a)23.96 (n/a)

layer_norm_1_cols_2_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.11 (+73.83%)0.09 (+50.07%)0.08 (+46.18%)0.06 (+7.28%)0.02 (+683.83%)138.10 (-6.82%)98.78 (-30.35%)97.10 (-31.57%)76.40 (-42.47%)24.88 (+313.86%)
0f0da0b — 2026-01-14 16:31:000.06 (n/a)0.06 (n/a)0.06 (n/a)0.06 (n/a)0.00 (n/a)148.20 (n/a)141.82 (n/a)141.90 (n/a)132.80 (n/a)6.01 (n/a)

layer_norm_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.12 (+58.00%)0.09 (+35.21%)0.09 (+33.88%)0.06 (+12.88%)0.02 (+142.40%)140.10 (-11.44%)98.38 (-23.32%)94.40 (-25.26%)68.10 (-36.71%)26.11 (+36.70%)
0f0da0b — 2026-01-14 16:31:000.08 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)158.20 (n/a)128.30 (n/a)126.30 (n/a)107.60 (n/a)19.10 (n/a)

layer_norm_2_cols_2_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.13 (+115.82%)0.09 (+79.97%)0.09 (+57.15%)0.06 (+84.94%)0.03 (+142.79%)130.40 (-45.94%)91.90 (-43.54%)93.50 (-36.39%)62.00 (-53.66%)25.59 (-42.10%)
0f0da0b — 2026-01-14 16:31:000.06 (n/a)0.05 (n/a)0.06 (n/a)0.03 (n/a)0.01 (n/a)241.20 (n/a)162.78 (n/a)147.00 (n/a)133.80 (n/a)44.20 (n/a)

layer_norm_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.11 (+63.95%)0.09 (+43.09%)0.11 (+55.71%)0.06 (-6.72%)0.02 (+806.48%)141.00 (+7.22%)91.54 (-25.97%)77.90 (-35.78%)73.00 (-39.07%)28.43 (+505.85%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.07 (n/a)0.07 (n/a)0.06 (n/a)0.00 (n/a)131.50 (n/a)123.66 (n/a)121.30 (n/a)119.80 (n/a)4.69 (n/a)

layer_norm_4_cols_2_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.11 (+76.69%)0.09 (+61.27%)0.09 (+70.26%)0.06 (+36.41%)0.02 (+195.74%)148.40 (-26.72%)103.14 (-34.74%)87.80 (-41.27%)73.70 (-43.44%)32.30 (+19.79%)
0f0da0b — 2026-01-14 16:31:000.06 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)202.50 (n/a)158.04 (n/a)149.50 (n/a)130.30 (n/a)26.96 (n/a)

layer_norm_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.12 (+75.09%)0.10 (+78.44%)0.11 (+94.76%)0.05 (+12.92%)0.03 (+213.89%)153.90 (-11.45%)90.36 (-39.81%)77.40 (-48.67%)67.30 (-42.87%)35.86 (+72.31%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.06 (n/a)0.05 (n/a)0.05 (n/a)0.01 (n/a)173.80 (n/a)150.12 (n/a)150.80 (n/a)117.80 (n/a)20.81 (n/a)

layer_norm_8_cols_2_channels_2048_tile_128

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.09 (+88.05%)0.08 (+72.40%)0.08 (+83.22%)0.06 (+32.78%)0.01 (+723.36%)140.30 (-24.69%)106.50 (-40.67%)97.60 (-45.44%)92.40 (-46.80%)19.84 (+234.33%)
0f0da0b — 2026-01-14 16:31:000.05 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.00 (n/a)186.30 (n/a)179.50 (n/a)178.90 (n/a)173.70 (n/a)5.93 (n/a)

matrix_vector_mul_128x128_32_1col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:490.37 (+17.13%)0.30 (+24.87%)0.30 (+38.03%)0.23 (+12.47%)0.06 (+35.23%)
0a6c11c — 2025-12-04 00:39:100.32 (n/a)0.24 (n/a)0.22 (n/a)0.20 (n/a)0.04 (n/a)

matrix_vector_mul_128x128_32_1col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:100.26 (-1.81%)0.24 (+4.06%)0.24 (+0.52%)0.21 (+16.97%)0.02 (-48.16%)0.26 (-1.81%)0.24 (+4.06%)0.24 (+0.52%)0.20 (+16.97%)0.02 (-48.16%)
a4b6ffe — 2026-01-05 19:24:050.26 (n/a)0.23 (n/a)0.24 (n/a)0.18 (n/a)0.04 (n/a)0.26 (n/a)0.23 (n/a)0.24 (n/a)0.17 (n/a)0.04 (n/a)

matrix_vector_mul_128x128_32tsi_128tso_1col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
1531781 — 2026-01-23 18:49:240.45 (+63.20%)0.35 (+43.38%)0.35 (+38.29%)0.22 (+12.34%)0.08 (+156.17%)0.44 (+63.20%)0.34 (+43.38%)0.34 (+38.29%)0.22 (+12.34%)0.08 (+156.17%)
0f0da0b — 2026-01-14 16:31:000.27 (n/a)0.24 (n/a)0.25 (n/a)0.20 (n/a)0.03 (n/a)0.27 (n/a)0.24 (n/a)0.25 (n/a)0.19 (n/a)0.03 (n/a)

matrix_vector_mul_2048x8192_1_1col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:4913.57 (+3.47%)13.15 (+1.09%)13.10 (+0.16%)12.52 (-1.58%)0.37 (+153.19%)
0a6c11c — 2025-12-04 00:39:1013.11 (n/a)13.01 (n/a)13.07 (n/a)12.72 (n/a)0.15 (n/a)

matrix_vector_mul_2048x8192_1_1col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:1013.43 (-0.07%)13.30 (+4.62%)13.27 (+1.29%)13.16 (+16.70%)0.12 (-86.69%)13.42 (-0.07%)13.29 (+4.62%)13.26 (+1.29%)13.16 (+16.70%)0.12 (-86.69%)
a4b6ffe — 2026-01-05 19:24:0513.44 (n/a)12.71 (n/a)13.10 (n/a)11.28 (n/a)0.87 (n/a)13.43 (n/a)12.70 (n/a)13.09 (n/a)11.27 (n/a)0.87 (n/a)

matrix_vector_mul_2048x8192_1_2col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:4925.06 (+2.47%)23.71 (+1.81%)24.13 (+0.50%)20.90 (+5.41%)1.48 (-15.60%)
0a6c11c — 2025-12-04 00:39:1024.46 (n/a)23.29 (n/a)24.01 (n/a)19.83 (n/a)1.75 (n/a)

matrix_vector_mul_2048x8192_1_2col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:1024.46 (-0.60%)21.07 (-8.03%)19.75 (-17.83%)18.08 (+1.18%)3.03 (+7.07%)24.44 (-0.60%)21.06 (-8.03%)19.74 (-17.83%)18.07 (+1.18%)3.03 (+7.07%)
a4b6ffe — 2026-01-05 19:24:0524.60 (n/a)22.91 (n/a)24.04 (n/a)17.87 (n/a)2.83 (n/a)24.59 (n/a)22.90 (n/a)24.02 (n/a)17.86 (n/a)2.83 (n/a)

matrix_vector_mul_2048x8192_1_4col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:4942.05 (+1.54%)40.50 (+12.60%)40.49 (+2.36%)38.98 (+97.53%)0.97 (-88.09%)
0a6c11c — 2025-12-04 00:39:1041.41 (n/a)35.97 (n/a)39.55 (n/a)19.73 (n/a)8.15 (n/a)

matrix_vector_mul_2048x8192_1_4col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:1042.58 (+1.51%)41.37 (+14.29%)41.25 (+4.89%)40.33 (+91.42%)0.92 (-89.18%)42.56 (+1.51%)41.34 (+14.29%)41.23 (+4.89%)40.30 (+91.42%)0.92 (-89.18%)
a4b6ffe — 2026-01-05 19:24:0541.95 (n/a)36.20 (n/a)39.33 (n/a)21.07 (n/a)8.54 (n/a)41.92 (n/a)36.17 (n/a)39.31 (n/a)21.05 (n/a)8.53 (n/a)

matrix_vector_mul_2048x8192_1_8col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:4946.62 (+3.11%)44.82 (+0.17%)44.78 (+0.17%)42.60 (-3.80%)1.60 (+367.72%)
0a6c11c — 2025-12-04 00:39:1045.21 (n/a)44.75 (n/a)44.70 (n/a)44.29 (n/a)0.34 (n/a)

matrix_vector_mul_2048x8192_1_8col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:1046.04 (+4.14%)45.06 (+3.47%)44.55 (+2.76%)44.30 (+2.84%)0.87 (+66.58%)46.01 (+4.14%)45.03 (+3.47%)44.52 (+2.76%)44.27 (+2.84%)0.87 (+66.58%)
a4b6ffe — 2026-01-05 19:24:0544.21 (n/a)43.55 (n/a)43.35 (n/a)43.07 (n/a)0.52 (n/a)44.18 (n/a)43.52 (n/a)43.32 (n/a)43.05 (n/a)0.52 (n/a)

matrix_vector_mul_2048x8192_1tsi_1024tso_2col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
1531781 — 2026-01-23 18:49:2424.80 (+0.16%)24.59 (-0.08%)24.55 (-0.22%)24.40 (+0.04%)0.18 (+11.75%)24.79 (+0.16%)24.58 (-0.08%)24.53 (-0.22%)24.38 (+0.04%)0.18 (+11.75%)
0f0da0b — 2026-01-14 16:31:0024.76 (n/a)24.61 (n/a)24.60 (n/a)24.39 (n/a)0.16 (n/a)24.75 (n/a)24.60 (n/a)24.58 (n/a)24.37 (n/a)0.16 (n/a)

matrix_vector_mul_2048x8192_1tsi_2048tso_1col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
1531781 — 2026-01-23 18:49:2413.54 (+0.77%)13.34 (+3.03%)13.45 (+0.75%)13.10 (+15.45%)0.21 (-77.02%)13.53 (+0.77%)13.33 (+3.03%)13.44 (+0.75%)13.09 (+15.45%)0.21 (-77.02%)
0f0da0b — 2026-01-14 16:31:0013.44 (n/a)12.95 (n/a)13.35 (n/a)11.35 (n/a)0.90 (n/a)13.43 (n/a)12.94 (n/a)13.34 (n/a)11.34 (n/a)0.90 (n/a)

matrix_vector_mul_2048x8192_1tsi_256tso_8col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
1531781 — 2026-01-23 18:49:2446.81 (+1.97%)45.91 (+2.21%)46.38 (+2.98%)44.81 (+2.36%)0.89 (-7.11%)46.79 (+1.97%)45.88 (+2.21%)46.35 (+2.98%)44.79 (+2.36%)0.89 (-7.11%)
0f0da0b — 2026-01-14 16:31:0045.91 (n/a)44.92 (n/a)45.04 (n/a)43.78 (n/a)0.96 (n/a)45.88 (n/a)44.89 (n/a)45.01 (n/a)43.75 (n/a)0.96 (n/a)

matrix_vector_mul_2048x8192_1tsi_512tso_4col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
1531781 — 2026-01-23 18:49:2441.49 (-0.92%)40.81 (+0.09%)41.13 (+0.38%)39.95 (+0.86%)0.65 (-24.40%)41.46 (-0.92%)40.79 (+0.09%)41.10 (+0.38%)39.93 (+0.86%)0.65 (-24.40%)
0f0da0b — 2026-01-14 16:31:0041.88 (n/a)40.78 (n/a)40.97 (n/a)39.61 (n/a)0.86 (n/a)41.85 (n/a)40.75 (n/a)40.95 (n/a)39.59 (n/a)0.86 (n/a)

matrix_vector_mul_8192x2048_4_1col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:4913.29 (+1.63%)13.19 (+6.08%)13.19 (+1.17%)13.09 (+26.86%)0.08 (-92.94%)
0a6c11c — 2025-12-04 00:39:1013.08 (n/a)12.44 (n/a)13.03 (n/a)10.32 (n/a)1.07 (n/a)

matrix_vector_mul_8192x2048_4_1col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:1013.68 (+3.42%)13.30 (+8.54%)13.38 (+7.25%)12.72 (+14.96%)0.35 (-66.09%)13.67 (+3.42%)13.29 (+8.54%)13.38 (+7.25%)12.71 (+14.96%)0.35 (-66.09%)
a4b6ffe — 2026-01-05 19:24:0513.23 (n/a)12.25 (n/a)12.48 (n/a)11.07 (n/a)1.04 (n/a)13.22 (n/a)12.24 (n/a)12.47 (n/a)11.06 (n/a)1.04 (n/a)

matrix_vector_mul_8192x2048_4_2col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:4925.58 (+4.24%)24.23 (+6.59%)24.03 (+0.29%)23.58 (+37.29%)0.70 (-74.97%)
0a6c11c — 2025-12-04 00:39:1024.54 (n/a)22.73 (n/a)23.96 (n/a)17.18 (n/a)2.79 (n/a)

matrix_vector_mul_8192x2048_4_2col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:1025.08 (+1.32%)24.68 (+13.42%)24.61 (+3.26%)24.47 (+48.32%)0.24 (-93.56%)25.07 (+1.32%)24.67 (+13.42%)24.59 (+3.26%)24.46 (+48.32%)0.24 (-93.56%)
a4b6ffe — 2026-01-05 19:24:0524.76 (n/a)21.76 (n/a)23.83 (n/a)16.50 (n/a)3.74 (n/a)24.74 (n/a)21.75 (n/a)23.82 (n/a)16.49 (n/a)3.73 (n/a)

matrix_vector_mul_8192x2048_4_4col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:4940.94 (-2.19%)39.74 (-1.10%)39.76 (-0.60%)38.17 (-2.14%)1.04 (+4.39%)
0a6c11c — 2025-12-04 00:39:1041.86 (n/a)40.18 (n/a)40.00 (n/a)39.00 (n/a)1.00 (n/a)

matrix_vector_mul_8192x2048_4_4col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:1040.74 (-1.09%)37.16 (-7.15%)40.25 (+1.47%)24.51 (-37.44%)7.07 (+736.13%)40.72 (-1.09%)37.13 (-7.15%)40.23 (+1.47%)24.50 (-37.44%)7.07 (+736.13%)
a4b6ffe — 2026-01-05 19:24:0541.19 (n/a)40.02 (n/a)39.67 (n/a)39.18 (n/a)0.85 (n/a)41.17 (n/a)39.99 (n/a)39.64 (n/a)39.16 (n/a)0.85 (n/a)

matrix_vector_mul_8192x2048_4_8col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:4946.71 (-1.04%)44.91 (-2.34%)44.71 (-2.98%)43.80 (-1.44%)0.96 (+6.93%)
0a6c11c — 2025-12-04 00:39:1047.20 (n/a)45.98 (n/a)46.08 (n/a)44.44 (n/a)0.90 (n/a)

matrix_vector_mul_8192x2048_4_8col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:1046.28 (+5.68%)44.89 (+14.59%)44.74 (+4.26%)44.20 (+62.80%)0.84 (-87.96%)46.25 (+5.68%)44.86 (+14.59%)44.71 (+4.26%)44.17 (+62.80%)0.84 (-87.96%)
a4b6ffe — 2026-01-05 19:24:0543.79 (n/a)39.17 (n/a)42.91 (n/a)27.15 (n/a)6.98 (n/a)43.77 (n/a)39.15 (n/a)42.88 (n/a)27.13 (n/a)6.98 (n/a)

matrix_vector_mul_8192x2048_4tsi_1024tso_1col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
1531781 — 2026-01-23 18:49:2413.46 (+0.53%)13.33 (+2.85%)13.32 (+1.17%)13.20 (+10.91%)0.11 (-81.73%)13.45 (+0.53%)13.32 (+2.85%)13.31 (+1.17%)13.19 (+10.91%)0.11 (-81.73%)
0f0da0b — 2026-01-14 16:31:0013.39 (n/a)12.96 (n/a)13.17 (n/a)11.90 (n/a)0.62 (n/a)13.38 (n/a)12.95 (n/a)13.16 (n/a)11.89 (n/a)0.62 (n/a)

matrix_vector_mul_8192x2048_4tsi_1024tso_2col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
1531781 — 2026-01-23 18:49:2425.27 (+0.55%)24.75 (-0.75%)24.61 (-1.28%)24.55 (-0.37%)0.30 (+52.22%)25.25 (+0.55%)24.74 (-0.75%)24.59 (-1.28%)24.53 (-0.37%)0.30 (+52.22%)
0f0da0b — 2026-01-14 16:31:0025.13 (n/a)24.94 (n/a)24.93 (n/a)24.64 (n/a)0.20 (n/a)25.11 (n/a)24.92 (n/a)24.91 (n/a)24.62 (n/a)0.20 (n/a)

matrix_vector_mul_8192x2048_4tsi_1024tso_4col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
1531781 — 2026-01-23 18:49:2440.86 (-4.37%)40.01 (-4.03%)40.43 (-3.32%)38.86 (-2.46%)0.89 (-24.84%)40.84 (-4.37%)39.98 (-4.03%)40.41 (-3.32%)38.84 (-2.46%)0.89 (-24.84%)
0f0da0b — 2026-01-14 16:31:0042.73 (n/a)41.69 (n/a)41.82 (n/a)39.85 (n/a)1.18 (n/a)42.70 (n/a)41.66 (n/a)41.80 (n/a)39.82 (n/a)1.18 (n/a)

matrix_vector_mul_8192x2048_4tsi_1024tso_8col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
1531781 — 2026-01-23 18:49:2446.08 (-1.82%)45.11 (-1.49%)45.49 (-0.25%)43.89 (-1.05%)0.97 (-6.46%)46.05 (-1.82%)45.08 (-1.49%)45.46 (-0.25%)43.87 (-1.05%)0.97 (-6.46%)
0f0da0b — 2026-01-14 16:31:0046.94 (n/a)45.79 (n/a)45.60 (n/a)44.36 (n/a)1.03 (n/a)46.91 (n/a)45.76 (n/a)45.58 (n/a)44.33 (n/a)1.03 (n/a)

mem_copy_16_cores_2_chans_2048_tile_128_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.07 (+22.03%)0.06 (+27.86%)0.06 (+35.49%)0.05 (+15.00%)0.01 (+49.69%)173.30 (-13.00%)137.94 (-21.32%)133.60 (-26.19%)123.40 (-18.06%)18.40 (+8.53%)
0a6c11c — 2025-12-04 00:39:100.05 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.00 (n/a)199.20 (n/a)175.32 (n/a)181.00 (n/a)150.60 (n/a)16.96 (n/a)

mem_copy_16_cores_2_chans_2048_tile_128_False0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.10 (+63.55%)0.08 (+85.69%)0.09 (+119.34%)0.04 (+23.15%)0.02 (+126.16%)195.30 (-18.79%)111.62 (-42.82%)86.60 (-54.42%)84.60 (-38.87%)47.65 (+9.68%)
0f0da0b — 2026-01-14 16:31:000.06 (n/a)0.04 (n/a)0.04 (n/a)0.03 (n/a)0.01 (n/a)240.50 (n/a)195.20 (n/a)190.00 (n/a)138.40 (n/a)43.45 (n/a)

mem_copy_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
003b49d — 2025-11-14 19:30:540.05 (-16.99%)0.05 (-12.47%)0.05 (-7.32%)0.04 (-11.17%)0.00 (-29.61%)197.00 (+12.57%)171.00 (+13.70%)164.00 (+7.89%)153.00 (+20.47%)17.49 (-4.32%)
0999a9a — 2025-11-13 19:24:080.06 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)175.00 (n/a)150.40 (n/a)152.00 (n/a)127.00 (n/a)18.28 (n/a)

mem_copy_1_cols_2_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
003b49d — 2025-11-14 19:30:540.06 (+5.56%)0.05 (-1.59%)0.05 (-1.39%)0.03 (-14.07%)0.01 (+25.03%)263.00 (+16.37%)166.00 (+4.53%)144.00 (+1.41%)126.00 (-5.26%)49.38 (+43.17%)
0999a9a — 2025-11-13 19:24:080.06 (n/a)0.05 (n/a)0.05 (n/a)0.03 (n/a)0.01 (n/a)226.00 (n/a)158.80 (n/a)142.00 (n/a)133.00 (n/a)34.49 (n/a)

mem_copy_1_cores_1_chans_2048_tile_2048_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.08 (+10.93%)0.07 (+7.79%)0.07 (+3.55%)0.05 (-1.73%)0.01 (+36.73%)150.20 (+1.76%)121.68 (-6.36%)125.60 (-3.38%)97.90 (-9.85%)18.56 (+21.98%)
0a6c11c — 2025-12-04 00:39:100.08 (n/a)0.06 (n/a)0.06 (n/a)0.06 (n/a)0.01 (n/a)147.60 (n/a)129.94 (n/a)130.00 (n/a)108.60 (n/a)15.21 (n/a)

mem_copy_1_cores_1_chans_2048_tile_2048_False0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.10 (+31.19%)0.09 (+26.03%)0.10 (+33.11%)0.07 (+25.01%)0.02 (+78.79%)118.10 (-20.04%)96.52 (-19.61%)84.40 (-24.91%)82.40 (-23.77%)18.29 (+9.09%)
0f0da0b — 2026-01-14 16:31:000.08 (n/a)0.07 (n/a)0.07 (n/a)0.06 (n/a)0.01 (n/a)147.70 (n/a)120.06 (n/a)112.40 (n/a)108.10 (n/a)16.76 (n/a)

mem_copy_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
003b49d — 2025-11-14 19:30:540.06 (-15.32%)0.05 (-17.09%)0.05 (-23.81%)0.04 (-15.54%)0.01 (-18.18%)193.00 (+18.40%)160.80 (+20.36%)168.00 (+31.25%)124.00 (+18.10%)22.92 (+11.39%)
0999a9a — 2025-11-13 19:24:080.07 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)163.00 (n/a)133.60 (n/a)128.00 (n/a)105.00 (n/a)20.58 (n/a)

mem_copy_2_cols_2_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
003b49d — 2025-11-14 19:30:540.06 (-23.02%)0.05 (-6.98%)0.05 (-2.13%)0.04 (-15.31%)0.01 (-33.29%)196.00 (+18.07%)147.00 (+6.52%)141.00 (+2.17%)126.00 (+29.90%)25.27 (+8.13%)
0999a9a — 2025-11-13 19:24:080.08 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)166.00 (n/a)138.00 (n/a)138.00 (n/a)97.00 (n/a)23.37 (n/a)

mem_copy_2_cores_1_chans_2048_tile_1024_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.10 (+31.37%)0.08 (+29.12%)0.08 (+29.06%)0.07 (+41.80%)0.01 (+0.23%)109.90 (-29.46%)98.30 (-23.38%)100.80 (-22.52%)79.40 (-23.87%)10.54 (-46.17%)
0a6c11c — 2025-12-04 00:39:100.08 (n/a)0.07 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)155.80 (n/a)128.30 (n/a)130.10 (n/a)104.30 (n/a)19.58 (n/a)

mem_copy_2_cores_1_chans_2048_tile_1024_False0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.12 (+103.55%)0.09 (+81.99%)0.09 (+69.86%)0.07 (+83.95%)0.02 (+187.42%)117.10 (-45.66%)92.20 (-43.53%)86.90 (-41.12%)70.40 (-50.84%)23.32 (-23.43%)
0f0da0b — 2026-01-14 16:31:000.06 (n/a)0.05 (n/a)0.06 (n/a)0.04 (n/a)0.01 (n/a)215.50 (n/a)163.26 (n/a)147.60 (n/a)143.20 (n/a)30.45 (n/a)

mem_copy_2_cores_2_chans_2048_tile_1024_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.10 (+25.02%)0.07 (+10.01%)0.07 (+7.53%)0.05 (-1.99%)0.01 (+77.08%)150.10 (+2.04%)114.60 (-7.22%)111.70 (-6.99%)83.90 (-20.02%)21.23 (+43.54%)
0a6c11c — 2025-12-04 00:39:100.08 (n/a)0.07 (n/a)0.07 (n/a)0.06 (n/a)0.01 (n/a)147.10 (n/a)123.52 (n/a)120.10 (n/a)104.90 (n/a)14.79 (n/a)

mem_copy_2_cores_2_chans_2048_tile_1024_False0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.10 (+35.25%)0.08 (+37.83%)0.09 (+48.91%)0.06 (+9.50%)0.02 (+125.91%)140.20 (-8.66%)101.66 (-25.39%)92.10 (-32.82%)82.90 (-26.11%)24.23 (+51.17%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)153.50 (n/a)136.26 (n/a)137.10 (n/a)112.20 (n/a)16.03 (n/a)

mem_copy_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
003b49d — 2025-11-14 19:30:540.07 (+11.97%)0.05 (+3.09%)0.05 (+2.10%)0.03 (-11.35%)0.01 (+41.65%)229.00 (+12.81%)156.00 (-0.51%)143.00 (-2.05%)117.00 (-10.69%)38.21 (+48.71%)
0999a9a — 2025-11-13 19:24:080.06 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)203.00 (n/a)156.80 (n/a)146.00 (n/a)131.00 (n/a)25.69 (n/a)

mem_copy_4_cols_2_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
003b49d — 2025-11-14 19:30:540.06 (+9.02%)0.05 (+5.59%)0.05 (+0.00%)0.04 (+3.63%)0.01 (+25.48%)193.00 (-3.50%)147.40 (-4.66%)147.00 (+0.00%)122.00 (-8.27%)25.24 (+7.48%)
0999a9a — 2025-11-13 19:24:080.06 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)200.00 (n/a)154.60 (n/a)147.00 (n/a)133.00 (n/a)23.48 (n/a)

mem_copy_4_cores_1_chans_2048_tile_512_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.09 (-3.98%)0.07 (+4.00%)0.07 (+7.04%)0.06 (+3.67%)0.01 (-15.76%)129.70 (-3.57%)113.30 (-4.50%)116.90 (-6.55%)90.40 (+4.15%)15.04 (-13.80%)
0a6c11c — 2025-12-04 00:39:100.09 (n/a)0.07 (n/a)0.07 (n/a)0.06 (n/a)0.01 (n/a)134.50 (n/a)118.64 (n/a)125.10 (n/a)86.80 (n/a)17.45 (n/a)

mem_copy_4_cores_1_chans_2048_tile_512_False0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.12 (+51.68%)0.10 (+55.58%)0.11 (+85.86%)0.06 (+8.92%)0.03 (+163.31%)128.30 (-8.23%)84.56 (-32.56%)71.60 (-46.17%)65.70 (-34.04%)26.42 (+57.86%)
0f0da0b — 2026-01-14 16:31:000.08 (n/a)0.07 (n/a)0.06 (n/a)0.06 (n/a)0.01 (n/a)139.80 (n/a)125.38 (n/a)133.00 (n/a)99.60 (n/a)16.73 (n/a)

mem_copy_4_cores_2_chans_2048_tile_512_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.08 (+9.57%)0.07 (+16.85%)0.07 (+19.58%)0.06 (+24.52%)0.01 (+1.05%)134.50 (-19.70%)116.64 (-14.92%)116.80 (-16.39%)98.00 (-8.75%)15.71 (-25.13%)
0a6c11c — 2025-12-04 00:39:100.08 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)167.50 (n/a)137.10 (n/a)139.70 (n/a)107.40 (n/a)20.99 (n/a)

mem_copy_4_cores_2_chans_2048_tile_512_False0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.10 (+21.52%)0.09 (+51.62%)0.10 (+79.69%)0.06 (+23.77%)0.02 (+25.48%)133.20 (-19.17%)92.98 (-33.88%)83.40 (-44.33%)82.10 (-17.74%)22.49 (-15.24%)
0f0da0b — 2026-01-14 16:31:000.08 (n/a)0.06 (n/a)0.05 (n/a)0.05 (n/a)0.01 (n/a)164.80 (n/a)140.62 (n/a)149.80 (n/a)99.80 (n/a)26.53 (n/a)

mem_copy_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
003b49d — 2025-11-14 19:30:540.05 (-13.48%)0.05 (-17.79%)0.05 (-15.33%)0.03 (-25.09%)0.01 (+6.67%)267.00 (+33.50%)176.80 (+24.51%)150.00 (+18.11%)141.00 (+15.57%)47.41 (+61.41%)
0999a9a — 2025-11-13 19:24:080.06 (n/a)0.06 (n/a)0.06 (n/a)0.04 (n/a)0.01 (n/a)200.00 (n/a)142.00 (n/a)127.00 (n/a)122.00 (n/a)29.37 (n/a)

mem_copy_8_cols_2_channels_2048_tile_128

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
003b49d — 2025-11-14 19:30:540.05 (+13.38%)0.05 (+23.85%)0.05 (+37.91%)0.05 (+30.72%)0.00 (-56.85%)166.00 (-23.50%)153.40 (-20.60%)153.00 (-27.49%)142.00 (-11.80%)7.63 (-70.60%)
0999a9a — 2025-11-13 19:24:080.05 (n/a)0.04 (n/a)0.04 (n/a)0.04 (n/a)0.01 (n/a)217.00 (n/a)193.20 (n/a)211.00 (n/a)161.00 (n/a)25.96 (n/a)

mem_copy_8_cores_1_chans_2048_tile_256_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.09 (+16.31%)0.08 (+15.34%)0.08 (+14.60%)0.07 (+19.30%)0.01 (-5.41%)117.30 (-16.15%)107.80 (-13.61%)108.50 (-12.78%)94.10 (-13.99%)8.29 (-32.01%)
0a6c11c — 2025-12-04 00:39:100.07 (n/a)0.07 (n/a)0.07 (n/a)0.06 (n/a)0.01 (n/a)139.90 (n/a)124.78 (n/a)124.40 (n/a)109.40 (n/a)12.20 (n/a)

mem_copy_8_cores_1_chans_2048_tile_256_False0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.09 (+46.69%)0.08 (+57.18%)0.09 (+64.02%)0.06 (+94.45%)0.01 (+6.58%)131.30 (-48.59%)99.78 (-38.77%)88.70 (-39.00%)86.90 (-31.84%)18.90 (-64.18%)
0f0da0b — 2026-01-14 16:31:000.06 (n/a)0.05 (n/a)0.06 (n/a)0.03 (n/a)0.01 (n/a)255.40 (n/a)162.96 (n/a)145.40 (n/a)127.50 (n/a)52.78 (n/a)

mem_copy_8_cores_2_chans_2048_tile_256_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.09 (+12.61%)0.07 (-2.24%)0.07 (-8.85%)0.06 (+5.13%)0.01 (+35.50%)136.50 (-4.88%)120.86 (+2.96%)125.10 (+9.64%)91.60 (-11.15%)15.72 (+10.46%)
0a6c11c — 2025-12-04 00:39:100.08 (n/a)0.07 (n/a)0.07 (n/a)0.06 (n/a)0.01 (n/a)143.50 (n/a)117.38 (n/a)114.10 (n/a)103.10 (n/a)14.23 (n/a)

mem_copy_8_cores_2_chans_2048_tile_256_False0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.10 (+63.97%)0.09 (+58.77%)0.09 (+53.52%)0.06 (+50.07%)0.02 (+70.34%)144.40 (-33.36%)97.70 (-36.49%)88.10 (-34.89%)80.00 (-39.02%)26.35 (-27.55%)
0f0da0b — 2026-01-14 16:31:000.06 (n/a)0.06 (n/a)0.06 (n/a)0.04 (n/a)0.01 (n/a)216.70 (n/a)153.84 (n/a)135.30 (n/a)131.20 (n/a)36.36 (n/a)

mha

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.18 (+0.16%)0.18 (+0.19%)0.18 (+0.12%)0.18 (+0.32%)0.00 (-52.24%)47253.90 (-0.32%)47213.70 (-0.19%)47210.90 (-0.12%)47167.70 (-0.16%)28.71 (-52.47%)
0a6c11c — 2025-12-04 00:39:100.18 (n/a)0.18 (n/a)0.18 (n/a)0.18 (n/a)0.00 (n/a)47405.00 (n/a)47301.72 (n/a)47265.90 (n/a)47241.80 (n/a)60.40 (n/a)

mha0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.18 (-0.08%)0.18 (+0.22%)0.18 (+0.32%)0.18 (+0.35%)0.00 (-42.58%)47367.50 (-0.34%)47232.68 (-0.22%)47232.00 (-0.32%)47073.00 (+0.08%)105.45 (-42.71%)
0f0da0b — 2026-01-14 16:31:000.18 (n/a)0.18 (n/a)0.18 (n/a)0.18 (n/a)0.00 (n/a)47531.20 (n/a)47335.18 (n/a)47384.30 (n/a)47037.20 (n/a)184.08 (n/a)

relu_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.11 (+69.81%)0.09 (+59.10%)0.10 (+75.49%)0.05 (+26.24%)0.03 (+206.46%)152.90 (-20.82%)101.62 (-32.81%)82.20 (-43.00%)73.00 (-41.13%)35.75 (+36.67%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.06 (n/a)0.06 (n/a)0.04 (n/a)0.01 (n/a)193.10 (n/a)151.24 (n/a)144.20 (n/a)124.00 (n/a)26.16 (n/a)

relu_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.10 (+60.88%)0.08 (+49.18%)0.09 (+55.13%)0.05 (+8.09%)0.02 (+316.89%)150.60 (-7.44%)101.26 (-30.06%)91.30 (-35.57%)83.00 (-37.87%)28.02 (+146.84%)
0f0da0b — 2026-01-14 16:31:000.06 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.00 (n/a)162.70 (n/a)144.78 (n/a)141.70 (n/a)133.60 (n/a)11.35 (n/a)

relu_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.10 (+36.44%)0.09 (+44.22%)0.09 (+49.00%)0.06 (+25.29%)0.02 (+65.69%)135.80 (-20.21%)96.20 (-29.75%)86.30 (-32.84%)83.10 (-26.72%)22.46 (-2.41%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)170.20 (n/a)136.94 (n/a)128.50 (n/a)113.40 (n/a)23.02 (n/a)

relu_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.11 (+58.81%)0.09 (+71.41%)0.10 (+74.46%)0.05 (+55.08%)0.02 (+79.07%)154.40 (-35.51%)93.32 (-40.67%)78.40 (-42.65%)75.80 (-37.04%)34.24 (-28.56%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.06 (n/a)0.06 (n/a)0.03 (n/a)0.01 (n/a)239.40 (n/a)157.28 (n/a)136.70 (n/a)120.40 (n/a)47.93 (n/a)

rms_norm_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.09 (+27.92%)0.08 (+35.45%)0.09 (+45.35%)0.06 (+20.87%)0.01 (+35.02%)131.50 (-17.30%)100.70 (-25.92%)92.40 (-31.20%)90.90 (-21.77%)17.35 (-10.74%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)159.00 (n/a)135.94 (n/a)134.30 (n/a)116.20 (n/a)19.43 (n/a)

rms_norm_1_cols_2_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.12 (+67.22%)0.10 (+57.83%)0.11 (+74.39%)0.06 (-3.10%)0.03 (+484.78%)145.20 (+3.20%)86.54 (-31.66%)72.10 (-42.64%)70.50 (-40.25%)32.80 (+266.62%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.06 (n/a)0.07 (n/a)0.06 (n/a)0.00 (n/a)140.70 (n/a)126.64 (n/a)125.70 (n/a)118.00 (n/a)8.95 (n/a)

rms_norm_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.09 (+26.55%)0.09 (+40.18%)0.09 (+47.01%)0.06 (+19.23%)0.01 (+54.24%)129.60 (-16.12%)98.22 (-28.12%)90.90 (-31.96%)89.80 (-20.95%)17.55 (+0.58%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)154.50 (n/a)136.64 (n/a)133.60 (n/a)113.60 (n/a)17.45 (n/a)

rms_norm_2_cols_2_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.13 (+67.61%)0.11 (+72.13%)0.11 (+66.55%)0.09 (+90.95%)0.02 (+33.97%)93.50 (-47.62%)74.32 (-42.73%)73.10 (-39.98%)62.40 (-40.34%)11.60 (-59.00%)
0f0da0b — 2026-01-14 16:31:000.08 (n/a)0.07 (n/a)0.07 (n/a)0.05 (n/a)0.01 (n/a)178.50 (n/a)129.78 (n/a)121.80 (n/a)104.60 (n/a)28.30 (n/a)

rms_norm_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.10 (+60.49%)0.09 (+76.01%)0.09 (+72.18%)0.06 (+64.97%)0.01 (+31.34%)130.70 (-39.41%)97.48 (-43.89%)90.20 (-41.92%)86.00 (-37.68%)18.66 (-50.69%)
0f0da0b — 2026-01-14 16:31:000.06 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)215.70 (n/a)173.74 (n/a)155.30 (n/a)138.00 (n/a)37.85 (n/a)

rms_norm_4_cols_2_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.11 (+60.28%)0.10 (+80.30%)0.11 (+105.91%)0.07 (+52.87%)0.02 (+90.94%)125.50 (-34.60%)83.90 (-43.66%)73.50 (-51.42%)72.90 (-37.64%)23.26 (-20.88%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.06 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)191.90 (n/a)148.92 (n/a)151.30 (n/a)116.90 (n/a)29.40 (n/a)

rms_norm_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.09 (+27.54%)0.08 (+37.27%)0.09 (+53.73%)0.05 (+11.36%)0.01 (+87.66%)154.30 (-10.24%)106.66 (-25.46%)94.80 (-34.98%)94.00 (-21.60%)26.64 (+33.21%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)171.90 (n/a)143.10 (n/a)145.80 (n/a)119.90 (n/a)20.00 (n/a)

rms_norm_8_cols_2_channels_2048_tile_128

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.09 (+38.93%)0.08 (+46.69%)0.09 (+73.48%)0.05 (+5.47%)0.02 (+140.19%)167.70 (-5.15%)109.36 (-28.83%)91.60 (-42.35%)88.90 (-28.02%)33.52 (+68.39%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.05 (n/a)0.05 (n/a)0.05 (n/a)0.01 (n/a)176.80 (n/a)153.66 (n/a)158.90 (n/a)123.50 (n/a)19.91 (n/a)

rope_1_cols_2_channels_4096_tile_4096_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1b2d230 — 2026-01-13 22:38:030.19 (-21.66%)0.17 (-8.53%)0.18 (+2.09%)0.14 (-6.71%)0.02 (-38.07%)174.10 (+7.20%)150.06 (+8.11%)139.60 (-2.04%)129.80 (+27.63%)20.36 (-10.65%)
d48746f — 2026-01-12 22:17:100.24 (n/a)0.18 (n/a)0.17 (n/a)0.15 (n/a)0.04 (n/a)162.40 (n/a)138.80 (n/a)142.50 (n/a)101.70 (n/a)22.79 (n/a)

rope_1c_32rows_512cols_32arows_0m

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:241.10 (+30.53%)0.99 (+42.49%)1.05 (+62.73%)0.73 (+21.76%)0.15 (+50.52%)133.80 (-17.86%)101.54 (-29.34%)93.30 (-38.58%)89.10 (-23.39%)18.60 (-3.62%)
0f0da0b — 2026-01-14 16:31:000.85 (n/a)0.69 (n/a)0.65 (n/a)0.60 (n/a)0.10 (n/a)162.90 (n/a)143.70 (n/a)151.90 (n/a)116.30 (n/a)19.30 (n/a)

rope_1c_32rows_512cols_8arows_0m

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.56 (-21.50%)0.52 (-8.95%)0.53 (-6.60%)0.48 (+7.05%)0.03 (-70.73%)153.70 (-6.57%)140.84 (+7.27%)140.10 (+7.03%)132.50 (+27.40%)8.40 (-65.00%)
0f0da0b — 2026-01-14 16:31:000.71 (n/a)0.58 (n/a)0.56 (n/a)0.45 (n/a)0.10 (n/a)164.50 (n/a)131.30 (n/a)130.90 (n/a)104.00 (n/a)24.00 (n/a)

rope_2_cols_2_channels_4096_tile_2048_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1b2d230 — 2026-01-13 22:38:030.21 (-0.56%)0.17 (-10.62%)0.16 (-13.58%)0.13 (-13.33%)0.03 (+17.05%)183.70 (+15.39%)151.76 (+12.77%)155.20 (+15.73%)117.30 (+0.60%)24.24 (+35.73%)
d48746f — 2026-01-12 22:17:100.21 (n/a)0.19 (n/a)0.18 (n/a)0.15 (n/a)0.02 (n/a)159.20 (n/a)134.58 (n/a)134.10 (n/a)116.60 (n/a)17.86 (n/a)

rope_2c_32rows_512cols_32arows_0m

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:241.25 (+43.66%)0.97 (+36.64%)0.91 (+36.32%)0.63 (+2.79%)0.25 (+150.75%)157.00 (-2.67%)108.10 (-23.42%)108.30 (-26.68%)78.70 (-30.42%)31.09 (+71.34%)
0f0da0b — 2026-01-14 16:31:000.87 (n/a)0.71 (n/a)0.67 (n/a)0.61 (n/a)0.10 (n/a)161.30 (n/a)141.16 (n/a)147.70 (n/a)113.10 (n/a)18.15 (n/a)

rope_2c_32rows_512cols_8arows_0m

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.63 (-10.72%)0.57 (+15.63%)0.57 (+20.61%)0.49 (+35.05%)0.05 (-60.20%)150.60 (-25.96%)130.58 (-17.08%)130.40 (-17.05%)116.80 (+12.09%)12.57 (-65.56%)
0f0da0b — 2026-01-14 16:31:000.71 (n/a)0.49 (n/a)0.47 (n/a)0.36 (n/a)0.13 (n/a)203.40 (n/a)157.48 (n/a)157.20 (n/a)104.20 (n/a)36.49 (n/a)

rope_4_cols_2_channels_4096_tile_1024_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1b2d230 — 2026-01-13 22:38:030.22 (+18.92%)0.17 (+6.24%)0.17 (+5.57%)0.14 (-0.50%)0.03 (+90.77%)171.10 (+0.47%)147.40 (-4.39%)148.10 (-5.31%)110.10 (-15.89%)23.59 (+60.67%)
d48746f — 2026-01-12 22:17:100.19 (n/a)0.16 (n/a)0.16 (n/a)0.14 (n/a)0.02 (n/a)170.30 (n/a)154.16 (n/a)156.40 (n/a)130.90 (n/a)14.69 (n/a)

rope_8_cols_2_channels_4096_tile_512_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1b2d230 — 2026-01-13 22:38:030.17 (-1.30%)0.15 (+6.63%)0.15 (-8.78%)0.14 (+39.68%)0.01 (-67.35%)174.90 (-28.41%)163.42 (-10.36%)167.40 (+9.63%)146.50 (+1.31%)11.19 (-76.17%)
d48746f — 2026-01-12 22:17:100.17 (n/a)0.14 (n/a)0.16 (n/a)0.10 (n/a)0.03 (n/a)244.30 (n/a)182.30 (n/a)152.70 (n/a)144.60 (n/a)46.98 (n/a)

rope_8c_32rows_512cols_32arows_0m

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:241.34 (+62.86%)0.89 (+47.92%)0.85 (+53.70%)0.60 (+49.43%)0.27 (+64.12%)163.50 (-33.07%)117.38 (-32.11%)115.50 (-34.93%)73.60 (-38.56%)32.08 (-33.78%)
0f0da0b — 2026-01-14 16:31:000.82 (n/a)0.60 (n/a)0.55 (n/a)0.40 (n/a)0.16 (n/a)244.30 (n/a)172.90 (n/a)177.50 (n/a)119.80 (n/a)48.44 (n/a)

rope_8c_32rows_512cols_8arows_0m

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.54 (-10.87%)0.52 (+6.74%)0.53 (+12.23%)0.48 (+19.79%)0.03 (-67.26%)152.90 (-16.49%)143.32 (-8.02%)139.70 (-10.91%)135.50 (+12.17%)7.46 (-68.88%)
0f0da0b — 2026-01-14 16:31:000.61 (n/a)0.48 (n/a)0.47 (n/a)0.40 (n/a)0.08 (n/a)183.10 (n/a)155.82 (n/a)156.80 (n/a)120.80 (n/a)23.99 (n/a)

rows_32-cols_512-angle_rows_32-aie_columns_1-method_type_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
82cf4ab — 2026-02-09 20:43:010.81 (n/a)0.64 (n/a)0.65 (n/a)0.47 (n/a)0.15 (n/a)208.90 (n/a)160.66 (n/a)150.20 (n/a)120.60 (n/a)38.66 (n/a)

rows_32-cols_512-angle_rows_32-aie_columns_2-method_type_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
82cf4ab — 2026-02-09 20:43:010.88 (n/a)0.63 (n/a)0.57 (n/a)0.45 (n/a)0.18 (n/a)217.40 (n/a)166.24 (n/a)171.90 (n/a)111.60 (n/a)43.60 (n/a)

rows_32-cols_512-angle_rows_32-aie_columns_8-method_type_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
82cf4ab — 2026-02-09 20:43:010.75 (n/a)0.58 (n/a)0.57 (n/a)0.41 (n/a)0.13 (n/a)241.80 (n/a)176.82 (n/a)173.30 (n/a)130.40 (n/a)41.37 (n/a)

rows_32-cols_512-angle_rows_8-aie_columns_1-method_type_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
82cf4ab — 2026-02-09 20:43:010.59 (n/a)0.49 (n/a)0.54 (n/a)0.34 (n/a)0.10 (n/a)220.10 (n/a)155.70 (n/a)135.90 (n/a)125.90 (n/a)38.69 (n/a)

rows_32-cols_512-angle_rows_8-aie_columns_2-method_type_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
82cf4ab — 2026-02-09 20:43:010.56 (n/a)0.46 (n/a)0.48 (n/a)0.31 (n/a)0.11 (n/a)236.00 (n/a)168.84 (n/a)152.80 (n/a)131.10 (n/a)43.97 (n/a)

rows_32-cols_512-angle_rows_8-aie_columns_8-method_type_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
82cf4ab — 2026-02-09 20:43:010.42 (n/a)0.32 (n/a)0.32 (n/a)0.21 (n/a)0.09 (n/a)357.40 (n/a)244.34 (n/a)232.70 (n/a)176.90 (n/a)72.40 (n/a)

seq_len_16384-dim_64-num_heads_1-num_pipelines_8

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
82cf4ab — 2026-02-09 20:43:010.18 (n/a)0.18 (n/a)0.18 (n/a)0.18 (n/a)0.00 (n/a)47682.50 (n/a)47504.22 (n/a)47493.20 (n/a)47189.70 (n/a)200.08 (n/a)

seq_len_256-embedding_dim_2048-hidden_dim_2048-prio_accuracy_False

No metrics available.

sigmoid_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.12 (+53.53%)0.10 (+54.49%)0.12 (+106.52%)0.05 (-0.16%)0.03 (+144.85%)153.20 (+0.13%)90.92 (-30.39%)68.60 (-51.55%)68.30 (-34.89%)36.80 (+58.03%)
0f0da0b — 2026-01-14 16:31:000.08 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)153.00 (n/a)130.62 (n/a)141.60 (n/a)104.90 (n/a)23.29 (n/a)

sigmoid_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.12 (+73.69%)0.10 (+60.60%)0.12 (+98.01%)0.06 (+2.86%)0.03 (+394.96%)140.80 (-2.83%)89.84 (-32.71%)68.90 (-49.52%)68.10 (-42.43%)32.04 (+163.16%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.06 (n/a)0.06 (n/a)0.06 (n/a)0.01 (n/a)144.90 (n/a)133.52 (n/a)136.50 (n/a)118.30 (n/a)12.18 (n/a)

sigmoid_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.10 (+30.15%)0.08 (+37.20%)0.09 (+57.41%)0.05 (+2.94%)0.02 (+110.00%)169.40 (-2.87%)106.58 (-23.47%)87.40 (-36.48%)84.00 (-23.15%)36.37 (+56.68%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)174.40 (n/a)139.26 (n/a)137.60 (n/a)109.30 (n/a)23.21 (n/a)

sigmoid_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.12 (+109.66%)0.10 (+84.35%)0.12 (+117.89%)0.05 (+3.89%)0.03 (+649.09%)157.30 (-3.73%)89.74 (-39.88%)68.70 (-54.11%)65.90 (-52.35%)38.85 (+251.55%)
0f0da0b — 2026-01-14 16:31:000.06 (n/a)0.06 (n/a)0.05 (n/a)0.05 (n/a)0.00 (n/a)163.40 (n/a)149.28 (n/a)149.70 (n/a)138.30 (n/a)11.05 (n/a)

silu_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.10 (+44.80%)0.08 (+33.22%)0.10 (+66.93%)0.06 (-1.08%)0.02 (+301.40%)143.30 (+1.06%)104.88 (-20.01%)82.20 (-40.13%)80.60 (-30.93%)32.24 (+177.47%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.06 (n/a)0.06 (n/a)0.06 (n/a)0.01 (n/a)141.80 (n/a)131.12 (n/a)137.30 (n/a)116.70 (n/a)11.62 (n/a)

silu_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.10 (+34.12%)0.08 (+47.27%)0.10 (+69.15%)0.06 (+39.46%)0.02 (+73.50%)138.90 (-28.29%)103.10 (-30.67%)83.20 (-40.91%)81.60 (-25.41%)28.30 (-8.88%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.06 (n/a)0.06 (n/a)0.04 (n/a)0.01 (n/a)193.70 (n/a)148.70 (n/a)140.80 (n/a)109.40 (n/a)31.06 (n/a)

silu_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.12 (+42.12%)0.11 (+61.07%)0.12 (+92.74%)0.06 (+8.38%)0.03 (+104.00%)137.40 (-7.72%)83.36 (-34.98%)70.00 (-48.07%)68.10 (-29.65%)30.27 (+32.68%)
0f0da0b — 2026-01-14 16:31:000.08 (n/a)0.07 (n/a)0.06 (n/a)0.06 (n/a)0.01 (n/a)148.90 (n/a)128.20 (n/a)134.80 (n/a)96.80 (n/a)22.82 (n/a)

silu_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.12 (+35.33%)0.09 (+59.58%)0.09 (+78.47%)0.07 (+60.20%)0.02 (-1.76%)109.80 (-37.58%)89.20 (-38.82%)87.90 (-43.98%)71.10 (-26.09%)14.50 (-53.17%)
0f0da0b — 2026-01-14 16:31:000.09 (n/a)0.06 (n/a)0.05 (n/a)0.05 (n/a)0.02 (n/a)175.90 (n/a)145.80 (n/a)156.90 (n/a)96.20 (n/a)30.97 (n/a)

softmax_1_cols_2_channels_4096_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.21 (+20.19%)0.16 (+13.84%)0.16 (+8.19%)0.12 (+26.61%)0.04 (+17.18%)135.50 (-20.99%)104.68 (-12.65%)102.70 (-7.56%)76.50 (-16.85%)22.55 (-26.21%)
0f0da0b — 2026-01-14 16:31:000.18 (n/a)0.14 (n/a)0.15 (n/a)0.10 (n/a)0.03 (n/a)171.50 (n/a)119.84 (n/a)111.10 (n/a)92.00 (n/a)30.56 (n/a)

softmax_2_cols_2_channels_4096_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.22 (+40.26%)0.18 (+43.33%)0.20 (+64.08%)0.13 (+24.34%)0.04 (+68.98%)127.80 (-19.62%)94.74 (-28.95%)83.40 (-39.08%)73.90 (-28.67%)24.46 (-5.31%)
0f0da0b — 2026-01-14 16:31:000.16 (n/a)0.13 (n/a)0.12 (n/a)0.10 (n/a)0.03 (n/a)159.00 (n/a)133.34 (n/a)136.90 (n/a)103.60 (n/a)25.83 (n/a)

softmax_2_cols_2_channels_4096_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.24 (+85.21%)0.17 (+69.97%)0.15 (+54.16%)0.11 (+37.21%)0.05 (+203.82%)148.10 (-27.12%)102.78 (-37.88%)106.70 (-35.10%)69.60 (-45.96%)31.81 (+15.98%)
0f0da0b — 2026-01-14 16:31:000.13 (n/a)0.10 (n/a)0.10 (n/a)0.08 (n/a)0.02 (n/a)203.20 (n/a)165.46 (n/a)164.40 (n/a)128.80 (n/a)27.43 (n/a)

swiglu

No metrics available.

swiglu_decode_1x2048x2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.00 (-15.46%)0.00 (-27.20%)0.00 (-28.17%)0.00 (-33.90%)0.00 (+1834.83%)5187.20 (+51.28%)4720.84 (+38.38%)4750.90 (+39.22%)4015.10 (+18.29%)390.23 (+3298.45%)
0a6c11c — 2025-12-04 00:39:100.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)3428.80 (n/a)3411.62 (n/a)3412.60 (n/a)3394.20 (n/a)11.48 (n/a)

swiglu_decode_1x2048x2048_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.00 (+28.14%)0.00 (+20.30%)0.00 (+10.00%)0.00 (+11.66%)0.00 (+124.73%)5111.40 (-10.44%)4324.88 (-14.91%)4736.10 (-9.09%)3410.70 (-21.96%)843.97 (+53.51%)
0f0da0b — 2026-01-14 16:31:000.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)5707.20 (n/a)5083.00 (n/a)5209.60 (n/a)4370.50 (n/a)549.78 (n/a)

tanh_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.10 (+42.65%)0.08 (+33.21%)0.07 (+27.20%)0.06 (+13.46%)0.02 (+130.64%)140.80 (-11.89%)109.14 (-23.00%)115.90 (-21.37%)85.60 (-29.89%)23.52 (+37.00%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)159.80 (n/a)141.74 (n/a)147.40 (n/a)122.10 (n/a)17.17 (n/a)

tanh_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.11 (+41.27%)0.09 (+40.30%)0.09 (+45.94%)0.08 (+28.08%)0.01 (+83.15%)103.30 (-21.98%)88.02 (-28.44%)87.80 (-31.46%)77.80 (-29.21%)9.64 (+2.40%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.07 (n/a)0.06 (n/a)0.06 (n/a)0.01 (n/a)132.40 (n/a)123.00 (n/a)128.10 (n/a)109.90 (n/a)9.41 (n/a)

tanh_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.10 (+48.49%)0.09 (+56.51%)0.09 (+68.30%)0.07 (+54.99%)0.01 (+44.32%)121.40 (-35.46%)93.40 (-36.24%)86.30 (-40.61%)84.60 (-32.70%)15.80 (-36.89%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.06 (n/a)0.06 (n/a)0.04 (n/a)0.01 (n/a)188.10 (n/a)146.48 (n/a)145.30 (n/a)125.70 (n/a)25.04 (n/a)

tanh_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.11 (+64.36%)0.09 (+41.14%)0.10 (+56.21%)0.05 (-7.33%)0.03 (+346.69%)163.80 (+7.91%)102.90 (-22.05%)80.40 (-35.94%)71.70 (-39.13%)39.99 (+185.25%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.06 (n/a)0.07 (n/a)0.05 (n/a)0.01 (n/a)151.80 (n/a)132.00 (n/a)125.50 (n/a)117.80 (n/a)14.02 (n/a)

transpose_2048_M_64_N_1_cols_1_channels_64_m_64_n_8_s

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:493.93 (-7.02%)3.68 (+2.48%)3.66 (+1.82%)3.45 (+15.62%)0.15 (-63.64%)151.80 (-13.50%)142.70 (-3.59%)143.40 (-1.78%)133.50 (+7.49%)5.90 (-66.22%)
0a6c11c — 2025-12-04 00:39:104.22 (n/a)3.59 (n/a)3.59 (n/a)2.99 (n/a)0.42 (n/a)175.50 (n/a)148.02 (n/a)146.00 (n/a)124.20 (n/a)17.46 (n/a)

transpose_2048_M_64_N_1_cols_1_channels_64_m_64_n_8_s0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:244.95 (+35.28%)4.34 (+33.32%)4.37 (+25.58%)3.74 (+35.73%)0.50 (+11.62%)140.30 (-26.35%)121.98 (-25.39%)119.90 (-20.33%)106.00 (-26.08%)14.15 (-39.67%)
0f0da0b — 2026-01-14 16:31:003.66 (n/a)3.26 (n/a)3.48 (n/a)2.75 (n/a)0.45 (n/a)190.50 (n/a)163.48 (n/a)150.50 (n/a)143.40 (n/a)23.46 (n/a)

transpose_2048_M_64_N_1_cols_2_channels_64_m_64_n_8_s

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:495.11 (+15.57%)3.85 (-3.32%)3.40 (-18.64%)3.29 (+7.09%)0.71 (+49.14%)159.60 (-6.61%)140.18 (+4.80%)154.20 (+22.97%)102.60 (-13.49%)22.70 (+19.77%)
0a6c11c — 2025-12-04 00:39:104.42 (n/a)3.99 (n/a)4.18 (n/a)3.07 (n/a)0.48 (n/a)170.90 (n/a)133.76 (n/a)125.40 (n/a)118.60 (n/a)18.95 (n/a)

transpose_2048_M_64_N_1_cols_2_channels_64_m_64_n_8_s0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:246.07 (+44.96%)4.99 (+45.77%)5.09 (+52.05%)4.09 (+35.80%)0.74 (+53.13%)128.10 (-26.34%)106.86 (-31.23%)103.00 (-34.27%)86.30 (-31.02%)15.67 (-22.94%)
0f0da0b — 2026-01-14 16:31:004.19 (n/a)3.42 (n/a)3.35 (n/a)3.01 (n/a)0.48 (n/a)173.90 (n/a)155.38 (n/a)156.70 (n/a)125.10 (n/a)20.34 (n/a)

weighted_rms_norm_1_cols_2_channels_2048_weights_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.11 (+41.27%)0.09 (+39.11%)0.09 (+47.90%)0.05 (+1.24%)0.02 (+132.19%)155.00 (-1.21%)99.92 (-24.84%)89.60 (-32.43%)76.50 (-29.17%)31.36 (+74.29%)
0f0da0b — 2026-01-14 16:31:000.08 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)156.90 (n/a)132.94 (n/a)132.60 (n/a)108.00 (n/a)17.99 (n/a)

weighted_rms_norm_2_cols_2_channels_2048_weights_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.10 (+48.28%)0.09 (+48.11%)0.09 (+60.52%)0.05 (+2.82%)0.02 (+134.54%)166.70 (-2.74%)102.72 (-28.96%)88.50 (-37.68%)83.10 (-32.55%)35.87 (+61.58%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)171.40 (n/a)144.60 (n/a)142.00 (n/a)123.20 (n/a)22.20 (n/a)

weighted_rms_norm_4_cols_2_channels_2048_weights_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.09 (+21.03%)0.09 (+31.88%)0.09 (+37.19%)0.07 (+25.91%)0.01 (+25.73%)119.00 (-20.61%)96.48 (-24.14%)89.90 (-27.09%)89.30 (-17.31%)12.81 (-17.86%)
0f0da0b — 2026-01-14 16:31:000.08 (n/a)0.07 (n/a)0.07 (n/a)0.05 (n/a)0.01 (n/a)149.90 (n/a)127.18 (n/a)123.30 (n/a)108.00 (n/a)15.59 (n/a)

weighted_rms_norm_8_cols_2_channels_2048_weights_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.11 (+67.16%)0.10 (+80.58%)0.10 (+80.44%)0.06 (+74.63%)0.02 (+72.16%)135.80 (-42.72%)89.92 (-44.57%)80.00 (-44.60%)74.30 (-40.23%)25.76 (-41.27%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.05 (n/a)0.06 (n/a)0.03 (n/a)0.01 (n/a)237.10 (n/a)162.22 (n/a)144.40 (n/a)124.30 (n/a)43.86 (n/a)

@github-actions
Copy link
Contributor

github-actions bot commented Feb 9, 2026

📊 Test Results for Test Example Applications

82cf4ab (2026_02_09_20_54_08)

IRONCLAD

Tested on 2026_02_09_20_54_08 at commit 82cf4ab.

Test Checks TTFT (mean)TPS (mean)
llama_3.2_1b_prompt_13_tokens_1 ❌ 0/1 n/a n/a
llama_3.2_1b_prompt_13_tokens_40 ❌ 0/1 n/a n/a
llama_3.2_1b_prompt_2048_tokens_1 ❌ 0/1 n/a n/a
llama_3.2_1b_prompt_2048_tokens_40 ❌ 0/1 n/a n/a
📈 Trends (vs main branch) for Test Example Applications

82cf4ab (2026_02_09_20_54_08)

IRONCLAD Trends

llama_3.2_1b

Commit/Date Num Tokens (max)Num Tokens (mean)Num Tokens (median)Num Tokens (min)Num Tokens (stddev)TPS (max)TPS (mean)TPS (median)TPS (min)TPS (stddev)TTFT (max)TTFT (mean)TTFT (median)TTFT (min)TTFT (stddev)Total (max)Total (mean)Total (median)Total (min)Total (stddev)
130b6ea — 2025-12-05 21:33:1240.00 (+0.00%)40.00 (+0.00%)40.00 (+0.00%)40.00 (+0.00%)0.00 (n/a)4.71 (-0.42%)4.64 (-0.09%)4.64 (+0.65%)4.55 (-0.22%)0.05 (-17.66%)4.41 (-0.34%)4.39 (-0.19%)4.38 (-0.33%)4.37 (-0.15%)0.01 (-25.90%)12.96 (-0.00%)12.80 (+0.07%)12.80 (-0.23%)12.67 (+0.44%)0.09 (-21.12%)
0a6c11c — 2025-12-03 23:35:1540.00 (n/a)40.00 (n/a)40.00 (n/a)40.00 (n/a)0.00 (n/a)4.73 (n/a)4.64 (n/a)4.61 (n/a)4.56 (n/a)0.06 (n/a)4.42 (n/a)4.40 (n/a)4.40 (n/a)4.37 (n/a)0.02 (n/a)12.96 (n/a)12.79 (n/a)12.83 (n/a)12.62 (n/a)0.12 (n/a)

llama_3.2_1b_prompt_13_tokens_1

Commit/Date Num_Tokens (max)Num_Tokens (mean)Num_Tokens (median)Num_Tokens (min)Num_Tokens (stddev)TPS (max)TPS (mean)TPS (median)TPS (min)TPS (stddev)TTFT (max)TTFT (mean)TTFT (median)TTFT (min)TTFT (stddev)
82cf4ab — 2026-02-09 20:53:37n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)
1531781 — 2026-01-23 18:40:341.00 (n/a)1.00 (n/a)1.00 (n/a)1.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.61 (n/a)0.61 (n/a)0.61 (n/a)0.61 (n/a)0.00 (n/a)

llama_3.2_1b_prompt_13_tokens_40

Commit/Date Num_Tokens (max)Num_Tokens (mean)Num_Tokens (median)Num_Tokens (min)Num_Tokens (stddev)TPS (max)TPS (mean)TPS (median)TPS (min)TPS (stddev)TTFT (max)TTFT (mean)TTFT (median)TTFT (min)TTFT (stddev)
82cf4ab — 2026-02-09 20:53:37n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)
1531781 — 2026-01-23 18:40:3440.00 (n/a)40.00 (n/a)40.00 (n/a)40.00 (n/a)0.00 (n/a)4.34 (n/a)4.34 (n/a)4.34 (n/a)4.34 (n/a)0.00 (n/a)0.61 (n/a)0.61 (n/a)0.61 (n/a)0.61 (n/a)0.00 (n/a)

llama_3.2_1b_prompt_2048_tokens_1

Commit/Date Num_Tokens (max)Num_Tokens (mean)Num_Tokens (median)Num_Tokens (min)Num_Tokens (stddev)TPS (max)TPS (mean)TPS (median)TPS (min)TPS (stddev)TTFT (max)TTFT (mean)TTFT (median)TTFT (min)TTFT (stddev)
82cf4ab — 2026-02-09 20:53:37n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)
1531781 — 2026-01-23 18:40:341.00 (n/a)1.00 (n/a)1.00 (n/a)1.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)2.70 (n/a)2.70 (n/a)2.70 (n/a)2.70 (n/a)0.00 (n/a)

llama_3.2_1b_prompt_2048_tokens_40

Commit/Date Num_Tokens (max)Num_Tokens (mean)Num_Tokens (median)Num_Tokens (min)Num_Tokens (stddev)TPS (max)TPS (mean)TPS (median)TPS (min)TPS (stddev)TTFT (max)TTFT (mean)TTFT (median)TTFT (min)TTFT (stddev)
82cf4ab — 2026-02-09 20:53:37n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)
1531781 — 2026-01-23 18:40:3440.00 (n/a)40.00 (n/a)40.00 (n/a)40.00 (n/a)0.00 (n/a)4.22 (n/a)4.22 (n/a)4.22 (n/a)4.22 (n/a)0.00 (n/a)2.70 (n/a)2.70 (n/a)2.70 (n/a)2.70 (n/a)0.00 (n/a)

@github-actions
Copy link
Contributor

github-actions bot commented Feb 9, 2026

📊 Test Results for Small Benchmark/Test Suite

a4c2152 (2026_02_09_21_09_35)

IRONCLAD

Tested on 2026_02_09_21_09_35 at commit a4c2152.

Test Checks Latency (mean)Bandwidth (mean)GFLOP/s (mean)
M_128-K_128-num_aie_columns_1-tile_size_input_32-tile_size_output_128 ✅ 5/5 n/a 0.28 n/a
M_1792-K_896-N_1152-num_aie_columns_8-b_col_maj_False-c_col_maj_True-m_64-k_32-n_48-trace_size_0-partition_N_1 ✅ 5/5 2456.44 3.87 n/a
M_192-K_384-N_64-num_aie_columns_4-b_col_maj_False-c_col_maj_False-m_48-k_96-n_16-trace_size_0-partition_N_1 ✅ 5/5 282.58 0.83 n/a
M_192-K_384-N_64-num_aie_columns_4-b_col_maj_True-c_col_maj_True-m_48-k_96-n_16-trace_size_0-partition_N_1 ✅ 5/5 264.12 0.93 n/a
M_2048-K_2048-N_2048-num_aie_columns_1-b_col_maj_False-c_col_maj_False-m_64-k_64-n_64-trace_size_0-partition_N_1 ✅ 5/5 48623.56 0.52 n/a
M_2048-K_2048-N_2048-num_aie_columns_2-b_col_maj_True-c_col_maj_False-m_64-k_64-n_64-trace_size_0-partition_N_1 ✅ 5/5 27855.38 0.90 n/a
M_2048-K_2048-N_2048-num_aie_columns_8-b_col_maj_True-c_col_maj_True-m_64-k_64-n_64-trace_size_0-partition_N_1 ✅ 5/5 7673.66 3.28 n/a
M_2048-K_8192-num_aie_columns_1-tile_size_input_1-tile_size_output_2048 ✅ 5/5 n/a 12.86 n/a
M_2048-K_8192-num_aie_columns_2-tile_size_input_1-tile_size_output_1024 ✅ 5/5 n/a 24.10 n/a
M_2048-K_8192-num_aie_columns_4-tile_size_input_1-tile_size_output_512 ✅ 5/5 n/a 40.97 n/a
M_2048-K_8192-num_aie_columns_8-tile_size_input_1-tile_size_output_256 ✅ 5/5 n/a 43.59 n/a
M_2048-N_64-aie_columns_1-channels_1-m_64-n_64-s_8 ✅ 5/5 173.60 3.18 n/a
M_2048-N_64-aie_columns_1-channels_2-m_64-n_64-s_8 ✅ 5/5 172.68 3.20 n/a
M_384-K_1536-N_1792-num_aie_columns_4-b_col_maj_True-c_col_maj_False-m_32-k_48-n_64-trace_size_0-partition_N_1 ✅ 5/5 2106.34 3.84 n/a
M_8192-K_2048-num_aie_columns_1-tile_size_input_4-tile_size_output_1024 ✅ 5/5 n/a 12.71 n/a
M_8192-K_2048-num_aie_columns_2-tile_size_input_4-tile_size_output_1024 ✅ 5/5 n/a 24.62 n/a
M_8192-K_2048-num_aie_columns_4-tile_size_input_4-tile_size_output_1024 ✅ 5/5 n/a 39.20 n/a
M_8192-K_2048-num_aie_columns_8-tile_size_input_4-tile_size_output_1024 ✅ 5/5 n/a 43.86 n/a
M_896-K_1792-N_640-num_aie_columns_8-b_col_maj_False-c_col_maj_True-m_32-k_64-n_80-trace_size_0-partition_N_1 ✅ 5/5 1331.04 5.06 n/a
embedding_dim_2048-hidden_dim_2048 ✅ 5/5 n/a n/a n/a
input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048 ✅ 30/30 153.13 0.06 n/a
input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048-group_size_32 ✅ 5/5 117.12 0.05 n/a
input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048-weighted_False ✅ 5/5 130.28 0.06 n/a
input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024 ✅ 10/10 135.14 0.06 n/a
input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024-group_size_32 ✅ 5/5 164.42 0.03 n/a
input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024-weighted_False ✅ 5/5 167.94 0.05 n/a
input_length_2048-num_aie_columns_1-num_channels_2-tile_size_2048 ✅ 10/10 137.55 0.09 n/a
input_length_2048-num_aie_columns_1-num_channels_2-tile_size_2048-scalar_factor_3.0 ✅ 5/5 151.88 0.08 n/a
input_length_2048-num_aie_columns_1-num_channels_2-tile_size_2048-weighted_True ✅ 5/5 154.48 0.08 n/a
input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024 ✅ 30/30 153.53 0.05 n/a
input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024-group_size_32 ✅ 5/5 163.36 0.03 n/a
input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024-weighted_False ✅ 5/5 147.60 0.06 n/a
input_length_2048-num_aie_columns_2-num_channels_2-tile_size_1024 ✅ 10/10 151.45 0.09 n/a
input_length_2048-num_aie_columns_2-num_channels_2-tile_size_1024-scalar_factor_3.0 ✅ 5/5 137.42 0.09 n/a
input_length_2048-num_aie_columns_2-num_channels_2-tile_size_1024-weighted_True ✅ 5/5 158.40 0.07 n/a
input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512 ✅ 10/10 163.41 0.05 n/a
input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512-group_size_32 ✅ 5/5 189.70 0.03 n/a
input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512-weighted_False ✅ 5/5 163.50 0.05 n/a
input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512 ✅ 30/30 164.10 0.05 n/a
input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512-group_size_32 ✅ 5/5 166.36 0.03 n/a
input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512-weighted_False ✅ 5/5 162.94 0.05 n/a
input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256 ✅ 10/10 174.86 0.05 n/a
input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256-group_size_32 ✅ 5/5 170.04 0.03 n/a
input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256-weighted_False ✅ 5/5 167.56 0.05 n/a
input_length_2048-num_aie_columns_4-num_channels_2-tile_size_512 ✅ 10/10 183.28 0.07 n/a
input_length_2048-num_aie_columns_4-num_channels_2-tile_size_512-scalar_factor_3.0 ✅ 5/5 187.88 0.07 n/a
input_length_2048-num_aie_columns_4-num_channels_2-tile_size_512-weighted_True ✅ 5/5 149.96 0.06 n/a
input_length_2048-num_aie_columns_8-num_channels_1-tile_size_256 ✅ 30/30 191.59 0.04 n/a
input_length_2048-num_aie_columns_8-num_channels_1-tile_size_256-group_size_32 ✅ 5/5 199.74 0.03 n/a
input_length_2048-num_aie_columns_8-num_channels_1-tile_size_256-weighted_False ✅ 5/5 186.36 0.04 n/a
input_length_2048-num_aie_columns_8-num_channels_2-tile_size_128 ✅ 10/10 199.24 0.04 n/a
input_length_2048-num_aie_columns_8-num_channels_2-tile_size_128-group_size_32 ✅ 5/5 205.60 0.03 n/a
input_length_2048-num_aie_columns_8-num_channels_2-tile_size_128-weighted_False ✅ 5/5 206.56 0.04 n/a
input_length_2048-num_aie_columns_8-num_channels_2-tile_size_256 ✅ 10/10 176.85 0.07 n/a
input_length_2048-num_aie_columns_8-num_channels_2-tile_size_256-scalar_factor_3.0 ✅ 5/5 284.10 0.05 n/a
input_length_2048-num_aie_columns_8-num_channels_2-tile_size_256-weighted_True ✅ 5/5 240.00 0.04 n/a
input_length_2048-num_cores_1-num_channels_1-bypass_False-tile_size_2048 ✅ 5/5 151.18 0.06 n/a
input_length_2048-num_cores_16-num_channels_2-bypass_False-tile_size_128 ✅ 5/5 202.82 0.04 n/a
input_length_2048-num_cores_2-num_channels_1-bypass_False-tile_size_1024 ✅ 5/5 136.02 0.06 n/a
input_length_2048-num_cores_2-num_channels_2-bypass_False-tile_size_1024 ✅ 5/5 146.76 0.06 n/a
input_length_2048-num_cores_4-num_channels_1-bypass_False-tile_size_512 ✅ 5/5 141.64 0.06 n/a
input_length_2048-num_cores_4-num_channels_2-bypass_False-tile_size_512 ✅ 5/5 151.28 0.06 n/a
input_length_2048-num_cores_8-num_channels_1-bypass_False-tile_size_256 ✅ 5/5 182.78 0.05 n/a
input_length_2048-num_cores_8-num_channels_2-bypass_False-tile_size_256 ✅ 5/5 143.54 0.06 n/a
input_length_32768-num_aie_columns_2-num_channels_2-tile_size_1024 ✅ 5/5 161.52 0.85 n/a
input_length_32768-num_aie_columns_2-num_channels_2-tile_size_512 ✅ 5/5 154.82 0.89 n/a
input_length_32768-num_aie_columns_4-num_channels_4-tile_size_2048 ✅ 5/5 198.06 0.67 n/a
rows_32-cols_512-angle_rows_32-aie_columns_1-method_type_0 ✅ 5/5 137.24 0.72 n/a
rows_32-cols_512-angle_rows_32-aie_columns_2-method_type_0 ✅ 5/5 134.14 0.75 n/a
rows_32-cols_512-angle_rows_32-aie_columns_8-method_type_0 ✅ 5/5 156.98 0.67 n/a
rows_32-cols_512-angle_rows_8-aie_columns_1-method_type_0 ✅ 5/5 173.24 0.46 n/a
rows_32-cols_512-angle_rows_8-aie_columns_2-method_type_0 ✅ 5/5 125.34 0.59 n/a
rows_32-cols_512-angle_rows_8-aie_columns_8-method_type_0 ✅ 5/5 165.10 0.45 n/a
seq_len_16384-dim_64-num_heads_1-num_pipelines_8 ✅ 5/5 47464.16 0.18 n/a
seq_len_256-embedding_dim_2048-hidden_dim_2048-prio_accuracy_False ❌ 0/5 n/a n/a n/a
📈 Trends (vs main branch) for Small Benchmark/Test Suite

a4c2152 (2026_02_09_21_09_35)

IRONCLAD Trends

M_128-K_128-num_aie_columns_1-tile_size_input_32-tile_size_output_128

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
a4c2152 — 2026-02-09 21:03:120.30 (n/a)0.28 (n/a)0.28 (n/a)0.25 (n/a)0.02 (n/a)0.29 (n/a)0.28 (n/a)0.28 (n/a)0.25 (n/a)0.02 (n/a)

M_1792-K_896-N_1152-num_aie_columns_8-b_col_maj_False-c_col_maj_True-m_64-k_32-n_48-trace_size_0-partition_N_1

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
a4c2152 — 2026-02-09 21:03:124.40 (n/a)3.87 (n/a)3.97 (n/a)3.22 (n/a)0.45 (n/a)2924.00 (n/a)2456.44 (n/a)2370.60 (n/a)2135.30 (n/a)300.27 (n/a)1732.50 (n/a)1523.10 (n/a)1560.53 (n/a)1265.19 (n/a)175.79 (n/a)

M_192-K_384-N_64-num_aie_columns_4-b_col_maj_False-c_col_maj_False-m_48-k_96-n_16-trace_size_0-partition_N_1

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
a4c2152 — 2026-02-09 21:03:121.06 (n/a)0.83 (n/a)0.84 (n/a)0.56 (n/a)0.20 (n/a)397.70 (n/a)282.58 (n/a)263.90 (n/a)209.60 (n/a)76.56 (n/a)45.03 (n/a)35.26 (n/a)35.76 (n/a)23.73 (n/a)8.71 (n/a)

M_192-K_384-N_64-num_aie_columns_4-b_col_maj_True-c_col_maj_True-m_48-k_96-n_16-trace_size_0-partition_N_1

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
a4c2152 — 2026-02-09 21:03:121.33 (n/a)0.93 (n/a)0.98 (n/a)0.50 (n/a)0.30 (n/a)440.60 (n/a)264.12 (n/a)225.10 (n/a)166.90 (n/a)104.55 (n/a)56.55 (n/a)39.55 (n/a)41.93 (n/a)21.42 (n/a)12.69 (n/a)

M_2048-K_2048-N_2048-num_aie_columns_1-b_col_maj_False-c_col_maj_False-m_64-k_64-n_64-trace_size_0-partition_N_1

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
a4c2152 — 2026-02-09 21:03:120.52 (n/a)0.52 (n/a)0.52 (n/a)0.52 (n/a)0.00 (n/a)48809.50 (n/a)48623.56 (n/a)48588.80 (n/a)48557.50 (n/a)104.94 (n/a)353.80 (n/a)353.33 (n/a)353.58 (n/a)351.98 (n/a)0.76 (n/a)

M_2048-K_2048-N_2048-num_aie_columns_2-b_col_maj_True-c_col_maj_False-m_64-k_64-n_64-trace_size_0-partition_N_1

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
a4c2152 — 2026-02-09 21:03:120.91 (n/a)0.90 (n/a)0.90 (n/a)0.90 (n/a)0.00 (n/a)27919.30 (n/a)27855.38 (n/a)27859.40 (n/a)27775.50 (n/a)65.40 (n/a)618.53 (n/a)616.75 (n/a)616.66 (n/a)615.34 (n/a)1.45 (n/a)

M_2048-K_2048-N_2048-num_aie_columns_8-b_col_maj_True-c_col_maj_True-m_64-k_64-n_64-trace_size_0-partition_N_1

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
a4c2152 — 2026-02-09 21:03:123.36 (n/a)3.28 (n/a)3.31 (n/a)3.18 (n/a)0.08 (n/a)7920.30 (n/a)7673.66 (n/a)7594.30 (n/a)7489.70 (n/a)197.75 (n/a)2293.80 (n/a)2240.00 (n/a)2262.19 (n/a)2169.10 (n/a)57.27 (n/a)

M_2048-K_8192-num_aie_columns_1-tile_size_input_1-tile_size_output_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
a4c2152 — 2026-02-09 21:03:1213.12 (n/a)12.86 (n/a)13.03 (n/a)12.23 (n/a)0.37 (n/a)13.11 (n/a)12.85 (n/a)13.02 (n/a)12.22 (n/a)0.37 (n/a)

M_2048-K_8192-num_aie_columns_2-tile_size_input_1-tile_size_output_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
a4c2152 — 2026-02-09 21:03:1224.72 (n/a)24.10 (n/a)23.91 (n/a)23.64 (n/a)0.48 (n/a)24.70 (n/a)24.08 (n/a)23.90 (n/a)23.63 (n/a)0.48 (n/a)

M_2048-K_8192-num_aie_columns_4-tile_size_input_1-tile_size_output_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
a4c2152 — 2026-02-09 21:03:1243.48 (n/a)40.97 (n/a)41.87 (n/a)38.54 (n/a)2.14 (n/a)43.45 (n/a)40.95 (n/a)41.85 (n/a)38.52 (n/a)2.13 (n/a)

M_2048-K_8192-num_aie_columns_8-tile_size_input_1-tile_size_output_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
a4c2152 — 2026-02-09 21:03:1246.58 (n/a)43.59 (n/a)42.95 (n/a)41.20 (n/a)2.03 (n/a)46.56 (n/a)43.57 (n/a)42.92 (n/a)41.18 (n/a)2.03 (n/a)

M_2048-N_64-aie_columns_1-channels_1-m_64-n_64-s_8

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
a4c2152 — 2026-02-09 21:03:123.97 (n/a)3.18 (n/a)3.47 (n/a)2.11 (n/a)0.75 (n/a)248.00 (n/a)173.60 (n/a)151.20 (n/a)132.10 (n/a)47.13 (n/a)

M_2048-N_64-aie_columns_1-channels_2-m_64-n_64-s_8

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
a4c2152 — 2026-02-09 21:03:124.05 (n/a)3.20 (n/a)3.14 (n/a)2.13 (n/a)0.76 (n/a)246.30 (n/a)172.68 (n/a)166.90 (n/a)129.50 (n/a)46.23 (n/a)

M_384-K_1536-N_1792-num_aie_columns_4-b_col_maj_True-c_col_maj_False-m_32-k_48-n_64-trace_size_0-partition_N_1

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
a4c2152 — 2026-02-09 21:03:124.17 (n/a)3.84 (n/a)3.89 (n/a)3.56 (n/a)0.28 (n/a)2263.30 (n/a)2106.34 (n/a)2072.30 (n/a)1931.00 (n/a)151.74 (n/a)1094.76 (n/a)1007.77 (n/a)1020.07 (n/a)934.00 (n/a)72.34 (n/a)

M_8192-K_2048-num_aie_columns_1-tile_size_input_4-tile_size_output_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
a4c2152 — 2026-02-09 21:03:1213.64 (n/a)12.71 (n/a)13.09 (n/a)11.73 (n/a)0.89 (n/a)13.63 (n/a)12.70 (n/a)13.08 (n/a)11.73 (n/a)0.89 (n/a)

M_8192-K_2048-num_aie_columns_2-tile_size_input_4-tile_size_output_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
a4c2152 — 2026-02-09 21:03:1225.58 (n/a)24.62 (n/a)24.78 (n/a)23.44 (n/a)0.80 (n/a)25.56 (n/a)24.60 (n/a)24.76 (n/a)23.43 (n/a)0.80 (n/a)

M_8192-K_2048-num_aie_columns_4-tile_size_input_4-tile_size_output_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
a4c2152 — 2026-02-09 21:03:1242.15 (n/a)39.20 (n/a)40.19 (n/a)36.15 (n/a)2.76 (n/a)42.12 (n/a)39.17 (n/a)40.16 (n/a)36.13 (n/a)2.76 (n/a)

M_8192-K_2048-num_aie_columns_8-tile_size_input_4-tile_size_output_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
a4c2152 — 2026-02-09 21:03:1246.13 (n/a)43.86 (n/a)44.12 (n/a)41.01 (n/a)2.05 (n/a)46.10 (n/a)43.84 (n/a)44.09 (n/a)40.98 (n/a)2.05 (n/a)

M_896-K_1792-N_640-num_aie_columns_8-b_col_maj_False-c_col_maj_True-m_32-k_64-n_80-trace_size_0-partition_N_1

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
a4c2152 — 2026-02-09 21:03:126.19 (n/a)5.06 (n/a)4.81 (n/a)4.49 (n/a)0.66 (n/a)1482.00 (n/a)1331.04 (n/a)1382.60 (n/a)1074.30 (n/a)154.26 (n/a)1913.04 (n/a)1562.94 (n/a)1486.49 (n/a)1386.77 (n/a)204.40 (n/a)

axpy_1_cols_2_channels_2048_tile_2048_3.0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.13 (-1.96%)0.11 (+2.52%)0.10 (+1.86%)0.08 (-1.91%)0.02 (-1.72%)162.40 (+1.95%)120.96 (-2.36%)117.00 (-1.85%)94.40 (+2.05%)23.11 (+4.09%)
0a6c11c — 2025-12-04 00:39:100.13 (n/a)0.10 (n/a)0.10 (n/a)0.08 (n/a)0.02 (n/a)159.30 (n/a)123.88 (n/a)119.20 (n/a)92.50 (n/a)22.21 (n/a)

axpy_1_cols_2_channels_2048_tile_2048_3.0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.16 (+80.10%)0.13 (+54.74%)0.11 (+34.32%)0.10 (+39.85%)0.03 (+432.10%)118.80 (-28.48%)100.30 (-33.22%)110.40 (-25.56%)78.00 (-44.48%)20.49 (+105.98%)
0f0da0b — 2026-01-14 16:31:000.09 (n/a)0.08 (n/a)0.08 (n/a)0.07 (n/a)0.01 (n/a)166.10 (n/a)150.20 (n/a)148.30 (n/a)140.50 (n/a)9.95 (n/a)

axpy_2_cols_2_channels_2048_tile_1024_3.0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.11 (-26.77%)0.09 (-9.09%)0.09 (-10.81%)0.08 (+8.25%)0.01 (-55.42%)152.00 (-7.60%)135.46 (+6.28%)142.80 (+12.18%)115.00 (+36.58%)15.03 (-41.36%)
0a6c11c — 2025-12-04 00:39:100.15 (n/a)0.10 (n/a)0.10 (n/a)0.07 (n/a)0.02 (n/a)164.50 (n/a)127.46 (n/a)127.30 (n/a)84.20 (n/a)25.64 (n/a)

axpy_2_cols_2_channels_2048_tile_1024_3.0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.14 (+54.25%)0.12 (+45.65%)0.14 (+61.31%)0.08 (-2.79%)0.03 (+557.87%)156.90 (+2.89%)104.46 (-28.09%)90.30 (-37.98%)87.90 (-35.22%)29.52 (+351.33%)
0f0da0b — 2026-01-14 16:31:000.09 (n/a)0.08 (n/a)0.08 (n/a)0.08 (n/a)0.00 (n/a)152.50 (n/a)145.26 (n/a)145.60 (n/a)135.70 (n/a)6.54 (n/a)

axpy_4_cols_2_channels_2048_tile_512_3.0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.12 (-10.21%)0.10 (-8.41%)0.10 (-6.20%)0.07 (+3.52%)0.02 (-23.48%)166.70 (-3.36%)128.78 (+7.33%)121.80 (+6.65%)103.90 (+11.36%)22.95 (-18.82%)
0a6c11c — 2025-12-04 00:39:100.13 (n/a)0.11 (n/a)0.11 (n/a)0.07 (n/a)0.02 (n/a)172.50 (n/a)119.98 (n/a)114.20 (n/a)93.30 (n/a)28.27 (n/a)

axpy_4_cols_2_channels_2048_tile_512_3.0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.16 (+65.19%)0.14 (+57.55%)0.15 (+63.48%)0.09 (+24.56%)0.03 (+220.28%)134.10 (-19.70%)92.44 (-34.51%)83.40 (-38.81%)78.60 (-39.49%)23.38 (+58.01%)
0f0da0b — 2026-01-14 16:31:000.09 (n/a)0.09 (n/a)0.09 (n/a)0.07 (n/a)0.01 (n/a)167.00 (n/a)141.16 (n/a)136.30 (n/a)129.90 (n/a)14.80 (n/a)

axpy_8_cols_2_channels_2048_tile_256_3.0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.10 (-16.19%)0.08 (-10.40%)0.09 (-2.09%)0.07 (-17.38%)0.01 (-10.61%)184.40 (+21.00%)153.18 (+11.99%)144.10 (+2.20%)127.10 (+19.34%)21.26 (+34.76%)
0a6c11c — 2025-12-04 00:39:100.12 (n/a)0.09 (n/a)0.09 (n/a)0.08 (n/a)0.01 (n/a)152.40 (n/a)136.78 (n/a)141.00 (n/a)106.50 (n/a)15.78 (n/a)

axpy_8_cols_2_channels_2048_tile_256_3.0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.17 (+67.03%)0.14 (+66.86%)0.15 (+90.72%)0.08 (+20.53%)0.03 (+162.97%)146.80 (-17.06%)92.24 (-37.11%)80.00 (-47.54%)72.30 (-40.15%)31.07 (+36.92%)
0f0da0b — 2026-01-14 16:31:000.10 (n/a)0.09 (n/a)0.08 (n/a)0.07 (n/a)0.01 (n/a)177.00 (n/a)146.68 (n/a)152.50 (n/a)120.80 (n/a)22.69 (n/a)

dequant_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.05 (-18.83%)0.04 (-5.21%)0.04 (-2.69%)0.03 (+37.10%)0.00 (-53.98%)160.40 (-27.06%)141.12 (-0.91%)136.40 (+2.79%)114.30 (+23.17%)17.37 (-59.06%)
0a6c11c — 2025-12-04 00:39:100.06 (n/a)0.04 (n/a)0.04 (n/a)0.02 (n/a)0.01 (n/a)219.90 (n/a)142.42 (n/a)132.70 (n/a)92.80 (n/a)42.43 (n/a)

dequant_1_cols_1_channels_2048_tile_2048_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.06 (+13.11%)0.05 (+1.55%)0.04 (-0.75%)0.04 (+7.05%)0.01 (+25.85%)125.50 (-6.55%)115.08 (-1.12%)121.60 (+0.75%)88.10 (-11.63%)15.40 (+4.09%)
0f0da0b — 2026-01-14 16:31:000.05 (n/a)0.05 (n/a)0.04 (n/a)0.04 (n/a)0.01 (n/a)134.30 (n/a)116.38 (n/a)120.70 (n/a)99.70 (n/a)14.79 (n/a)

dequant_1_cols_2_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.06 (-13.28%)0.05 (-8.93%)0.06 (-13.59%)0.04 (+4.63%)0.01 (-39.10%)133.60 (-4.43%)106.08 (+5.85%)95.10 (+15.69%)87.90 (+15.35%)17.90 (-31.97%)
0a6c11c — 2025-12-04 00:39:100.07 (n/a)0.06 (n/a)0.06 (n/a)0.04 (n/a)0.01 (n/a)139.80 (n/a)100.22 (n/a)82.20 (n/a)76.20 (n/a)26.31 (n/a)

dequant_1_cols_2_channels_2048_tile_1024_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.07 (+24.45%)0.05 (+9.74%)0.05 (-2.94%)0.03 (+16.88%)0.01 (+14.24%)163.70 (-14.47%)114.66 (-9.57%)111.30 (+3.06%)76.50 (-19.64%)31.63 (-21.04%)
0f0da0b — 2026-01-14 16:31:000.06 (n/a)0.04 (n/a)0.05 (n/a)0.03 (n/a)0.01 (n/a)191.40 (n/a)126.80 (n/a)108.00 (n/a)95.20 (n/a)40.06 (n/a)

dequant_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.07 (+5.66%)0.05 (+1.21%)0.05 (+0.61%)0.04 (-15.02%)0.01 (+58.42%)143.40 (+17.64%)101.02 (+1.49%)97.60 (-0.61%)78.90 (-5.40%)22.56 (+78.52%)
0a6c11c — 2025-12-04 00:39:100.06 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)121.90 (n/a)99.54 (n/a)98.20 (n/a)83.40 (n/a)12.64 (n/a)

dequant_2_cols_1_channels_2048_tile_1024_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.05 (+2.33%)0.05 (+22.41%)0.05 (+32.19%)0.04 (+7.46%)0.01 (-3.28%)147.80 (-6.93%)113.14 (-18.51%)106.50 (-24.31%)101.40 (-2.31%)19.61 (-8.79%)
0f0da0b — 2026-01-14 16:31:000.05 (n/a)0.04 (n/a)0.04 (n/a)0.03 (n/a)0.01 (n/a)158.80 (n/a)138.84 (n/a)140.70 (n/a)103.80 (n/a)21.51 (n/a)

dequant_2_cols_2_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.05 (+17.06%)0.04 (+8.98%)0.05 (+21.37%)0.03 (-11.65%)0.01 (+301.89%)159.70 (+13.18%)126.32 (-5.68%)109.00 (-17.67%)106.80 (-14.56%)22.48 (+280.15%)
0a6c11c — 2025-12-04 00:39:100.04 (n/a)0.04 (n/a)0.04 (n/a)0.04 (n/a)0.00 (n/a)141.10 (n/a)133.92 (n/a)132.40 (n/a)125.00 (n/a)5.91 (n/a)

dequant_2_cols_2_channels_2048_tile_512_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.05 (+14.83%)0.04 (+10.04%)0.05 (+21.96%)0.03 (-12.01%)0.01 (+143.96%)158.30 (+13.64%)121.38 (-6.49%)106.20 (-17.99%)98.20 (-12.94%)26.13 (+141.52%)
0f0da0b — 2026-01-14 16:31:000.05 (n/a)0.04 (n/a)0.04 (n/a)0.04 (n/a)0.00 (n/a)139.30 (n/a)129.80 (n/a)129.50 (n/a)112.80 (n/a)10.82 (n/a)

dequant_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.06 (+20.24%)0.05 (-3.19%)0.05 (-6.73%)0.03 (-20.69%)0.01 (+194.26%)156.10 (+26.09%)120.86 (+7.15%)114.90 (+7.18%)88.30 (-16.85%)24.40 (+214.82%)
0a6c11c — 2025-12-04 00:39:100.05 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.00 (n/a)123.80 (n/a)112.80 (n/a)107.20 (n/a)106.20 (n/a)7.75 (n/a)

dequant_4_cols_1_channels_2048_tile_512_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.06 (+28.85%)0.05 (+16.79%)0.05 (+23.60%)0.03 (-12.66%)0.01 (+157.83%)164.80 (+14.52%)113.62 (-10.20%)103.20 (-19.12%)84.80 (-22.42%)32.55 (+130.11%)
0f0da0b — 2026-01-14 16:31:000.05 (n/a)0.04 (n/a)0.04 (n/a)0.04 (n/a)0.00 (n/a)143.90 (n/a)126.52 (n/a)127.60 (n/a)109.30 (n/a)14.15 (n/a)

dequant_4_cols_2_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.05 (-6.09%)0.04 (-8.04%)0.04 (-12.46%)0.03 (-0.41%)0.01 (-20.19%)168.10 (+0.42%)135.74 (+7.44%)132.20 (+14.26%)107.40 (+6.44%)22.91 (-12.67%)
0a6c11c — 2025-12-04 00:39:100.05 (n/a)0.04 (n/a)0.05 (n/a)0.03 (n/a)0.01 (n/a)167.40 (n/a)126.34 (n/a)115.70 (n/a)100.90 (n/a)26.23 (n/a)

dequant_4_cols_2_channels_2048_tile_256_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.07 (+42.90%)0.05 (+60.94%)0.06 (+84.55%)0.03 (+50.15%)0.01 (+27.03%)188.40 (-33.40%)112.72 (-39.08%)93.80 (-45.84%)79.70 (-29.96%)43.58 (-36.67%)
0f0da0b — 2026-01-14 16:31:000.05 (n/a)0.03 (n/a)0.03 (n/a)0.02 (n/a)0.01 (n/a)282.90 (n/a)185.04 (n/a)173.20 (n/a)113.80 (n/a)68.82 (n/a)

dequant_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.04 (-13.65%)0.04 (-13.67%)0.04 (-18.56%)0.03 (-13.52%)0.00 (-23.88%)170.40 (+15.68%)140.12 (+15.33%)135.80 (+22.78%)119.60 (+15.78%)17.94 (+1.87%)
0a6c11c — 2025-12-04 00:39:100.05 (n/a)0.04 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)147.30 (n/a)121.50 (n/a)110.60 (n/a)103.30 (n/a)17.61 (n/a)

dequant_8_cols_1_channels_2048_tile_256_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.04 (-5.92%)0.04 (-11.07%)0.04 (-12.74%)0.03 (-9.17%)0.00 (-4.85%)166.90 (+10.09%)138.66 (+12.51%)135.90 (+14.59%)117.60 (+6.33%)17.94 (+10.10%)
0f0da0b — 2026-01-14 16:31:000.05 (n/a)0.04 (n/a)0.04 (n/a)0.03 (n/a)0.00 (n/a)151.60 (n/a)123.24 (n/a)118.60 (n/a)110.60 (n/a)16.30 (n/a)

dequant_8_cols_2_channels_2048_tile_128

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.05 (+17.21%)0.03 (-1.70%)0.03 (-3.04%)0.03 (-16.20%)0.01 (+95.59%)198.20 (+19.33%)160.00 (+4.92%)166.20 (+3.17%)106.50 (-14.66%)29.86 (+87.19%)
0a6c11c — 2025-12-04 00:39:100.04 (n/a)0.03 (n/a)0.03 (n/a)0.03 (n/a)0.00 (n/a)166.10 (n/a)152.50 (n/a)161.10 (n/a)124.80 (n/a)15.95 (n/a)

dequant_8_cols_2_channels_2048_tile_128_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.05 (+10.58%)0.04 (+11.15%)0.04 (+32.75%)0.02 (-15.30%)0.01 (+23.92%)229.90 (+18.08%)145.60 (-7.34%)131.00 (-24.67%)106.20 (-9.54%)49.03 (+42.68%)
0f0da0b — 2026-01-14 16:31:000.04 (n/a)0.03 (n/a)0.03 (n/a)0.03 (n/a)0.01 (n/a)194.70 (n/a)157.14 (n/a)173.90 (n/a)117.40 (n/a)34.37 (n/a)

eltwise_add_1_cols_2_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.17 (+59.76%)0.12 (+36.17%)0.10 (+5.69%)0.08 (+17.86%)0.04 (+129.37%)155.80 (-15.19%)109.78 (-22.78%)120.50 (-5.34%)73.80 (-37.40%)34.38 (+16.64%)
0f0da0b — 2026-01-14 16:31:000.10 (n/a)0.09 (n/a)0.10 (n/a)0.07 (n/a)0.02 (n/a)183.70 (n/a)142.16 (n/a)127.30 (n/a)117.90 (n/a)29.48 (n/a)

eltwise_add_2_cols_2_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.15 (+24.10%)0.13 (+23.91%)0.13 (+30.32%)0.08 (+16.80%)0.03 (+31.09%)146.70 (-14.36%)101.20 (-18.68%)91.70 (-23.26%)81.40 (-19.41%)26.00 (-6.71%)
0f0da0b — 2026-01-14 16:31:000.12 (n/a)0.10 (n/a)0.10 (n/a)0.07 (n/a)0.02 (n/a)171.30 (n/a)124.44 (n/a)119.50 (n/a)101.00 (n/a)27.87 (n/a)

eltwise_add_4_cols_2_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.17 (+52.81%)0.15 (+77.83%)0.15 (+81.17%)0.11 (+106.70%)0.02 (+14.06%)116.00 (-51.65%)86.46 (-45.77%)80.90 (-44.82%)72.70 (-34.56%)17.22 (-64.39%)
0f0da0b — 2026-01-14 16:31:000.11 (n/a)0.08 (n/a)0.08 (n/a)0.05 (n/a)0.02 (n/a)239.90 (n/a)159.44 (n/a)146.60 (n/a)111.10 (n/a)48.37 (n/a)

eltwise_add_8_cols_2_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.16 (+58.53%)0.14 (+73.02%)0.14 (+70.95%)0.11 (+188.94%)0.02 (-23.24%)109.50 (-65.38%)90.00 (-48.14%)86.00 (-41.50%)75.30 (-36.93%)13.44 (-83.58%)
0f0da0b — 2026-01-14 16:31:000.10 (n/a)0.08 (n/a)0.08 (n/a)0.04 (n/a)0.03 (n/a)316.30 (n/a)173.56 (n/a)147.00 (n/a)119.40 (n/a)81.81 (n/a)

eltwise_mul_1_cols_2_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.14 (+42.19%)0.13 (+64.04%)0.13 (+74.51%)0.10 (+75.35%)0.01 (+0.96%)118.10 (-42.97%)97.88 (-39.87%)92.60 (-42.70%)90.40 (-29.70%)11.55 (-59.47%)
0f0da0b — 2026-01-14 16:31:000.10 (n/a)0.08 (n/a)0.08 (n/a)0.06 (n/a)0.01 (n/a)207.10 (n/a)162.78 (n/a)161.60 (n/a)128.60 (n/a)28.50 (n/a)

eltwise_mul_2_cols_2_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.13 (+44.68%)0.12 (+43.52%)0.13 (+47.71%)0.09 (+35.56%)0.02 (+80.76%)130.00 (-26.26%)102.10 (-29.85%)93.30 (-32.34%)91.10 (-30.88%)16.53 (-8.94%)
0f0da0b — 2026-01-14 16:31:000.09 (n/a)0.09 (n/a)0.09 (n/a)0.07 (n/a)0.01 (n/a)176.30 (n/a)145.54 (n/a)137.90 (n/a)131.80 (n/a)18.15 (n/a)

eltwise_mul_4_cols_2_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.17 (+63.62%)0.15 (+73.87%)0.15 (+72.82%)0.11 (+104.66%)0.02 (+10.66%)114.00 (-51.16%)86.18 (-44.46%)81.50 (-42.12%)74.20 (-38.88%)15.85 (-65.75%)
0f0da0b — 2026-01-14 16:31:000.10 (n/a)0.08 (n/a)0.09 (n/a)0.05 (n/a)0.02 (n/a)233.40 (n/a)155.18 (n/a)140.80 (n/a)121.40 (n/a)46.29 (n/a)

eltwise_mul_8_cols_2_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.13 (+46.12%)0.11 (+59.06%)0.12 (+66.26%)0.08 (+48.55%)0.02 (+55.56%)154.80 (-32.70%)112.28 (-36.89%)104.60 (-39.85%)95.50 (-31.59%)24.14 (-26.88%)
0f0da0b — 2026-01-14 16:31:000.09 (n/a)0.07 (n/a)0.07 (n/a)0.05 (n/a)0.01 (n/a)230.00 (n/a)177.92 (n/a)173.90 (n/a)139.60 (n/a)33.01 (n/a)

embedding_dim_2048-hidden_dim_2048

No metrics available.

gelu_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.11 (+53.19%)0.10 (+56.21%)0.11 (+82.01%)0.08 (+32.44%)0.02 (+220.41%)101.60 (-24.52%)83.80 (-34.61%)72.40 (-45.11%)72.20 (-34.72%)15.75 (+58.78%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.06 (n/a)0.06 (n/a)0.06 (n/a)0.01 (n/a)134.60 (n/a)128.16 (n/a)131.90 (n/a)110.60 (n/a)9.92 (n/a)

gelu_1_cols_2_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.12 (+63.09%)0.11 (+76.43%)0.12 (+100.14%)0.07 (+62.57%)0.02 (+62.52%)110.00 (-38.48%)80.26 (-43.34%)69.10 (-50.00%)66.40 (-38.69%)18.54 (-39.34%)
0f0da0b — 2026-01-14 16:31:000.08 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)178.80 (n/a)141.66 (n/a)138.20 (n/a)108.30 (n/a)30.55 (n/a)

gelu_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.12 (+51.06%)0.10 (+80.20%)0.12 (+120.49%)0.06 (+32.94%)0.03 (+101.39%)137.80 (-24.74%)85.02 (-42.29%)68.20 (-54.65%)67.80 (-33.79%)30.24 (+4.82%)
0f0da0b — 2026-01-14 16:31:000.08 (n/a)0.06 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)183.10 (n/a)147.32 (n/a)150.40 (n/a)102.40 (n/a)28.85 (n/a)

gelu_2_cols_2_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.12 (+49.86%)0.11 (+81.45%)0.12 (+78.91%)0.08 (+119.08%)0.02 (+8.63%)107.80 (-54.38%)76.80 (-47.80%)69.50 (-44.13%)67.10 (-33.23%)17.40 (-67.51%)
0f0da0b — 2026-01-14 16:31:000.08 (n/a)0.06 (n/a)0.07 (n/a)0.03 (n/a)0.02 (n/a)236.30 (n/a)147.12 (n/a)124.40 (n/a)100.50 (n/a)53.56 (n/a)

gelu_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.13 (+73.67%)0.11 (+92.11%)0.12 (+121.10%)0.05 (+47.52%)0.03 (+98.64%)153.10 (-32.20%)85.56 (-45.77%)69.30 (-54.77%)63.00 (-42.47%)37.94 (-17.18%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.06 (n/a)0.05 (n/a)0.04 (n/a)0.02 (n/a)225.80 (n/a)157.78 (n/a)153.20 (n/a)109.50 (n/a)45.81 (n/a)

gelu_4_cols_2_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.12 (+87.02%)0.11 (+117.65%)0.12 (+144.58%)0.07 (+103.45%)0.02 (+63.92%)110.50 (-50.82%)79.80 (-54.61%)70.50 (-59.08%)70.00 (-46.56%)17.48 (-56.79%)
0f0da0b — 2026-01-14 16:31:000.06 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)224.70 (n/a)175.80 (n/a)172.30 (n/a)131.00 (n/a)40.46 (n/a)

gelu_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.12 (+61.36%)0.10 (+65.53%)0.11 (+79.09%)0.04 (-5.63%)0.03 (+170.92%)208.20 (+5.95%)98.74 (-30.26%)71.90 (-44.13%)70.20 (-38.04%)61.20 (+79.92%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.06 (n/a)0.06 (n/a)0.04 (n/a)0.01 (n/a)196.50 (n/a)141.58 (n/a)128.70 (n/a)113.30 (n/a)34.01 (n/a)

gelu_8_cols_2_channels_2048_tile_128

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.10 (+64.20%)0.09 (+72.61%)0.10 (+77.79%)0.06 (+39.91%)0.02 (+119.55%)133.40 (-28.55%)93.44 (-40.99%)82.30 (-43.75%)81.90 (-39.11%)22.47 (-5.49%)
0f0da0b — 2026-01-14 16:31:000.06 (n/a)0.05 (n/a)0.06 (n/a)0.04 (n/a)0.01 (n/a)186.70 (n/a)158.34 (n/a)146.30 (n/a)134.50 (n/a)23.77 (n/a)

gemm_1792x896x1152_64x32x48_8cols_ccolmaj

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
1531781 — 2026-01-23 18:49:245.15 (+4.03%)4.90 (+9.86%)4.99 (+12.67%)4.27 (+4.36%)0.36 (+11.38%)2201.70 (-4.17%)1929.94 (-8.91%)1885.00 (-11.25%)1826.70 (-3.87%)154.05 (+4.53%)2025.19 (+4.03%)1925.76 (+9.86%)1962.51 (+12.67%)1680.20 (+4.36%)139.97 (+11.38%)
0f0da0b — 2026-01-14 16:31:004.95 (n/a)4.46 (n/a)4.43 (n/a)4.09 (n/a)0.32 (n/a)2297.60 (n/a)2118.80 (n/a)2123.90 (n/a)1900.20 (n/a)147.37 (n/a)1946.79 (n/a)1752.94 (n/a)1741.75 (n/a)1610.08 (n/a)125.67 (n/a)

gemm_192x384x64_48x96x16_4cols

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
1531781 — 2026-01-23 18:49:242.36 (+52.61%)1.85 (+39.10%)1.78 (+41.59%)1.32 (+14.31%)0.39 (+132.56%)167.80 (-12.51%)124.24 (-26.22%)124.40 (-29.36%)93.70 (-34.52%)28.11 (+35.82%)100.67 (+52.61%)78.94 (+39.10%)75.88 (+41.59%)56.25 (+14.31%)16.73 (+132.56%)
0f0da0b — 2026-01-14 16:31:001.55 (n/a)1.33 (n/a)1.26 (n/a)1.15 (n/a)0.17 (n/a)191.80 (n/a)168.40 (n/a)176.10 (n/a)143.10 (n/a)20.70 (n/a)65.96 (n/a)56.75 (n/a)53.59 (n/a)49.20 (n/a)7.20 (n/a)

gemm_192x384x64_48x96x16_4cols_bcolmaj_ccolmaj

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
1531781 — 2026-01-23 18:49:242.22 (+30.89%)2.06 (+54.21%)2.20 (+75.49%)1.60 (+35.74%)0.26 (+25.57%)138.20 (-26.33%)109.20 (-35.27%)100.50 (-43.03%)99.80 (-23.58%)16.54 (-28.10%)94.54 (+30.89%)87.77 (+54.21%)93.91 (+75.49%)68.30 (+35.74%)11.26 (+25.57%)
0f0da0b — 2026-01-14 16:31:001.69 (n/a)1.33 (n/a)1.25 (n/a)1.18 (n/a)0.21 (n/a)187.60 (n/a)168.70 (n/a)176.40 (n/a)130.60 (n/a)23.00 (n/a)72.23 (n/a)56.92 (n/a)53.51 (n/a)50.31 (n/a)8.97 (n/a)

gemm_2048x2048x2048_64x64x32_8_cols_0_bcolmaj_0_ccolmaj_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)GFLOP/s (max)GFLOP/s (mean)GFLOP/s (median)GFLOP/s (min)GFLOP/s (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3a053d3 — 2025-11-26 01:12:163.72 (-0.11%)3.55 (-2.75%)3.56 (-2.75%)3.36 (-5.33%)0.12 (+76.10%)2725.23 (-0.11%)2602.52 (-2.75%)2612.11 (-2.75%)2459.89 (-5.33%)86.27 (+76.10%)6984.00 (+5.63%)6608.60 (+2.91%)6577.00 (+2.83%)6304.00 (+0.11%)221.87 (+87.15%)
9e98049 — 2025-11-25 20:50:553.72 (n/a)3.65 (n/a)3.66 (n/a)3.54 (n/a)0.07 (n/a)2728.26 (n/a)2676.06 (n/a)2686.03 (n/a)2598.29 (n/a)48.99 (n/a)6612.00 (n/a)6422.00 (n/a)6396.00 (n/a)6297.00 (n/a)118.55 (n/a)

gemm_2048x2048x2048_64x64x32_8_cols_0_bcolmaj_1_ccolmaj_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)GFLOP/s (max)GFLOP/s (mean)GFLOP/s (median)GFLOP/s (min)GFLOP/s (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3a053d3 — 2025-11-26 01:12:163.40 (-3.14%)3.38 (-0.82%)3.38 (-0.82%)3.35 (+4.93%)0.02 (-84.51%)2493.81 (-3.14%)2475.27 (-0.82%)2479.06 (-0.82%)2453.92 (+4.93%)13.33 (-84.51%)7001.00 (-4.70%)6940.80 (+0.71%)6930.00 (+0.83%)6889.00 (+3.24%)37.43 (-84.76%)
9e98049 — 2025-11-25 20:50:553.51 (n/a)3.40 (n/a)3.41 (n/a)3.19 (n/a)0.12 (n/a)2574.53 (n/a)2495.72 (n/a)2499.62 (n/a)2338.67 (n/a)86.02 (n/a)7346.00 (n/a)6892.20 (n/a)6873.00 (n/a)6673.00 (n/a)245.57 (n/a)

gemm_2048x2048x2048_64x64x32_8_cols_1_bcolmaj_0_ccolmaj_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)GFLOP/s (max)GFLOP/s (mean)GFLOP/s (median)GFLOP/s (min)GFLOP/s (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3a053d3 — 2025-11-26 01:12:163.02 (+0.52%)2.87 (-2.05%)2.86 (-2.24%)2.70 (-6.16%)0.11 (+131.66%)2216.19 (+0.52%)2104.44 (-2.05%)2094.08 (-2.24%)1980.84 (-6.16%)80.25 (+131.67%)8673.00 (+6.56%)8175.60 (+2.21%)8204.00 (+2.29%)7752.00 (-0.51%)314.03 (+145.36%)
9e98049 — 2025-11-25 20:50:553.01 (n/a)2.93 (n/a)2.92 (n/a)2.88 (n/a)0.05 (n/a)2204.81 (n/a)2148.41 (n/a)2142.13 (n/a)2110.81 (n/a)34.64 (n/a)8139.00 (n/a)7998.60 (n/a)8020.00 (n/a)7792.00 (n/a)127.99 (n/a)

gemm_2048x2048x2048_64x64x64_1cols

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
1531781 — 2026-01-23 18:49:240.52 (-0.75%)0.52 (-0.07%)0.52 (+0.01%)0.52 (+0.22%)0.00 (-86.34%)48415.10 (-0.22%)48370.52 (+0.07%)48362.20 (-0.01%)48344.40 (+0.76%)29.40 (-86.25%)355.36 (-0.75%)355.17 (-0.07%)355.23 (+0.01%)354.85 (+0.22%)0.22 (-86.34%)
0f0da0b — 2026-01-14 16:31:000.52 (n/a)0.52 (n/a)0.52 (n/a)0.52 (n/a)0.00 (n/a)48520.80 (n/a)48335.84 (n/a)48365.40 (n/a)47981.00 (n/a)213.79 (n/a)358.06 (n/a)355.43 (n/a)355.21 (n/a)354.07 (n/a)1.58 (n/a)

gemm_2048x2048x2048_64x64x64_2_cols_0_bcolmaj_0_ccolmaj_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:491.04 (+0.94%)1.03 (+0.23%)1.03 (+0.09%)1.02 (+0.07%)0.00 (+197.51%)24576.00 (-0.07%)24485.82 (-0.23%)24530.40 (-0.09%)24266.40 (-0.94%)111.26 (+194.35%)
0a6c11c — 2025-12-04 00:39:101.03 (n/a)1.03 (n/a)1.03 (n/a)1.02 (n/a)0.00 (n/a)24594.30 (n/a)24542.14 (n/a)24551.90 (n/a)24495.50 (n/a)37.80 (n/a)

gemm_2048x2048x2048_64x64x64_2_cols_0_bcolmaj_0_ccolmaj_0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
a4b6ffe — 2026-01-05 19:24:051.04 (+0.28%)1.03 (-0.15%)1.03 (+0.05%)1.02 (-0.52%)0.01 (+42.72%)24719.80 (+0.52%)24512.42 (+0.16%)24529.20 (-0.05%)24092.60 (-0.28%)252.92 (+43.15%)713.08 (+0.28%)700.92 (-0.15%)700.38 (+0.05%)694.98 (-0.52%)7.30 (+42.71%)
50ad37c — 2025-12-29 20:35:531.04 (n/a)1.03 (n/a)1.03 (n/a)1.02 (n/a)0.01 (n/a)24592.30 (n/a)24473.70 (n/a)24540.90 (n/a)24160.80 (n/a)176.68 (n/a)711.06 (n/a)702.00 (n/a)700.05 (n/a)698.59 (n/a)5.12 (n/a)

gemm_2048x2048x2048_64x64x64_2_cols_0_bcolmaj_1_ccolmaj_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.99 (+1.69%)0.97 (+0.29%)0.97 (-0.04%)0.97 (-0.15%)0.01 (+434.92%)26025.30 (+0.15%)25839.34 (-0.29%)25916.70 (+0.04%)25446.00 (-1.67%)202.05 (+425.44%)
0a6c11c — 2025-12-04 00:39:100.97 (n/a)0.97 (n/a)0.97 (n/a)0.97 (n/a)0.00 (n/a)25986.90 (n/a)25913.52 (n/a)25905.70 (n/a)25877.00 (n/a)38.45 (n/a)

gemm_2048x2048x2048_64x64x64_2_cols_0_bcolmaj_1_ccolmaj_0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
a4b6ffe — 2026-01-05 19:24:050.97 (-0.01%)0.97 (+0.03%)0.97 (+0.04%)0.97 (+0.00%)0.00 (-6.43%)25927.90 (-0.00%)25903.20 (-0.03%)25898.60 (-0.04%)25892.80 (+0.01%)14.05 (-6.56%)663.50 (-0.01%)663.23 (+0.03%)663.35 (+0.04%)662.60 (+0.00%)0.36 (-6.43%)
50ad37c — 2025-12-29 20:35:530.97 (n/a)0.97 (n/a)0.97 (n/a)0.97 (n/a)0.00 (n/a)25928.10 (n/a)25910.14 (n/a)25908.10 (n/a)25891.20 (n/a)15.04 (n/a)663.54 (n/a)663.06 (n/a)663.11 (n/a)662.60 (n/a)0.38 (n/a)

gemm_2048x2048x2048_64x64x64_2_cols_1_bcolmaj_0_ccolmaj_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.92 (-1.02%)0.92 (-0.66%)0.92 (-0.10%)0.91 (-0.78%)0.01 (-4.54%)27751.40 (+0.79%)27480.74 (+0.66%)27354.80 (+0.10%)27320.50 (+1.03%)182.69 (-2.84%)
0a6c11c — 2025-12-04 00:39:100.93 (n/a)0.92 (n/a)0.92 (n/a)0.91 (n/a)0.01 (n/a)27534.80 (n/a)27299.32 (n/a)27326.70 (n/a)27042.30 (n/a)188.04 (n/a)

gemm_2048x2048x2048_64x64x64_2_cols_1_bcolmaj_0_ccolmaj_0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
a4b6ffe — 2026-01-05 19:24:050.92 (+0.11%)0.91 (-0.12%)0.91 (-0.21%)0.91 (-0.24%)0.00 (+30.34%)27723.50 (+0.24%)27548.94 (+0.12%)27522.90 (+0.21%)27403.60 (-0.11%)124.26 (+30.58%)626.92 (+0.11%)623.62 (-0.12%)624.20 (-0.21%)619.69 (-0.24%)2.81 (+30.34%)
50ad37c — 2025-12-29 20:35:530.92 (n/a)0.91 (n/a)0.92 (n/a)0.91 (n/a)0.00 (n/a)27657.10 (n/a)27515.58 (n/a)27464.70 (n/a)27434.60 (n/a)95.16 (n/a)626.21 (n/a)624.37 (n/a)625.52 (n/a)621.17 (n/a)2.16 (n/a)

gemm_2048x2048x2048_64x64x64_2cols_bcolmaj

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
1531781 — 2026-01-23 18:49:240.93 (+1.26%)0.91 (+0.29%)0.91 (+0.24%)0.90 (-0.34%)0.01 (+132.37%)27826.70 (+0.34%)27533.84 (-0.28%)27519.90 (-0.24%)27105.20 (-1.24%)272.35 (+129.91%)633.82 (+1.26%)624.00 (+0.29%)624.27 (+0.24%)617.39 (-0.34%)6.21 (+132.36%)
0f0da0b — 2026-01-14 16:31:000.92 (n/a)0.91 (n/a)0.91 (n/a)0.91 (n/a)0.00 (n/a)27731.40 (n/a)27612.36 (n/a)27585.70 (n/a)27446.50 (n/a)118.46 (n/a)625.94 (n/a)622.19 (n/a)622.78 (n/a)619.51 (n/a)2.67 (n/a)

gemm_2048x2048x2048_64x64x64_8_cols_0_bcolmaj_0_ccolmaj_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)GFLOP/s (max)GFLOP/s (mean)GFLOP/s (median)GFLOP/s (min)GFLOP/s (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
31ab24e — 2025-11-22 02:01:563.65 (-1.06%)3.59 (+0.83%)3.63 (-0.26%)3.45 (+2.66%)0.07 (-44.71%)2678.08 (-1.06%)2632.13 (+0.83%)2658.60 (-0.26%)2528.31 (+2.66%)53.86 (-44.71%)6795.00 (-2.59%)6529.80 (-0.92%)6462.00 (+0.26%)6415.00 (+1.07%)137.14 (-45.27%)
d9a64c5 — 2025-11-21 20:49:353.69 (n/a)3.56 (n/a)3.64 (n/a)3.36 (n/a)0.13 (n/a)2706.77 (n/a)2610.50 (n/a)2665.61 (n/a)2462.71 (n/a)97.41 (n/a)6976.00 (n/a)6590.40 (n/a)6445.00 (n/a)6347.00 (n/a)250.58 (n/a)

gemm_2048x2048x2048_64x64x64_8_cols_0_bcolmaj_0_ccolmaj_0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
d48746f — 2026-01-12 22:17:103.98 (+0.20%)3.84 (-2.95%)3.76 (-5.27%)3.72 (-4.97%)0.13 (+473.97%)6760.00 (+5.23%)6559.50 (+3.13%)6697.10 (+5.56%)6321.00 (-0.20%)218.68 (+499.40%)2717.91 (+0.20%)2621.44 (-2.95%)2565.27 (-5.27%)2541.40 (-4.97%)88.37 (+473.97%)
331dcca — 2026-01-08 18:34:103.97 (n/a)3.96 (n/a)3.97 (n/a)3.92 (n/a)0.02 (n/a)6423.80 (n/a)6360.32 (n/a)6344.30 (n/a)6333.90 (n/a)36.48 (n/a)2712.39 (n/a)2701.17 (n/a)2707.91 (n/a)2674.40 (n/a)15.40 (n/a)

gemm_2048x2048x2048_64x64x64_8_cols_0_bcolmaj_1_ccolmaj_0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
d48746f — 2026-01-12 22:17:103.75 (+0.28%)3.70 (+1.25%)3.71 (-0.21%)3.65 (+4.77%)0.04 (-66.90%)6901.80 (-4.56%)6795.46 (-1.30%)6790.20 (+0.21%)6715.90 (-0.28%)67.05 (-68.40%)2558.08 (+0.28%)2528.33 (+1.25%)2530.10 (-0.21%)2489.18 (+4.77%)24.81 (-66.90%)
331dcca — 2026-01-08 18:34:103.74 (n/a)3.66 (n/a)3.71 (n/a)3.48 (n/a)0.11 (n/a)7231.20 (n/a)6885.18 (n/a)6775.70 (n/a)6734.90 (n/a)212.21 (n/a)2550.87 (n/a)2497.04 (n/a)2535.50 (n/a)2375.78 (n/a)74.96 (n/a)

gemm_2048x2048x2048_64x64x64_8_cols_1_bcolmaj_0_ccolmaj_0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
d48746f — 2026-01-12 22:17:103.62 (-0.77%)3.54 (+0.14%)3.60 (-0.04%)3.44 (+2.17%)0.09 (-25.77%)7307.40 (-2.13%)7103.46 (-0.18%)6982.50 (+0.04%)6948.20 (+0.77%)182.95 (-26.55%)2472.55 (-0.77%)2419.80 (+0.14%)2460.41 (-0.04%)2351.04 (+2.17%)61.76 (-25.77%)
331dcca — 2026-01-08 18:34:103.65 (n/a)3.54 (n/a)3.61 (n/a)3.37 (n/a)0.12 (n/a)7466.20 (n/a)7116.40 (n/a)6979.60 (n/a)6895.10 (n/a)249.09 (n/a)2491.61 (n/a)2416.45 (n/a)2461.45 (n/a)2301.00 (n/a)83.20 (n/a)

gemm_2048x2048x2048_64x64x64_8cols_bcolmaj_ccolmaj

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
1531781 — 2026-01-23 18:49:243.46 (+1.18%)3.42 (+2.60%)3.45 (+2.90%)3.29 (+2.92%)0.07 (-25.59%)7651.20 (-2.83%)7368.34 (-2.57%)7300.80 (-2.82%)7263.70 (-1.17%)162.53 (-28.22%)2365.17 (+1.18%)2332.46 (+2.60%)2353.14 (+2.90%)2245.39 (+2.92%)50.15 (-25.59%)
0f0da0b — 2026-01-14 16:31:003.42 (n/a)3.33 (n/a)3.35 (n/a)3.20 (n/a)0.10 (n/a)7874.40 (n/a)7562.58 (n/a)7512.80 (n/a)7349.50 (n/a)226.42 (n/a)2337.56 (n/a)2273.31 (n/a)2286.73 (n/a)2181.74 (n/a)67.40 (n/a)

gemm_384x1536x1792_32x48x64_4cols_bcolmaj

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
1531781 — 2026-01-23 18:49:244.35 (-0.80%)4.24 (+3.81%)4.33 (+9.58%)3.85 (-0.98%)0.22 (-4.90%)2094.90 (+0.99%)1905.64 (-3.69%)1862.00 (-8.75%)1852.00 (+0.81%)105.88 (-3.06%)1141.45 (-0.80%)1111.86 (+3.81%)1135.29 (+9.58%)1009.08 (-0.98%)57.51 (-4.90%)
0f0da0b — 2026-01-14 16:31:004.39 (n/a)4.08 (n/a)3.95 (n/a)3.89 (n/a)0.23 (n/a)2074.30 (n/a)1978.70 (n/a)2040.50 (n/a)1837.10 (n/a)109.22 (n/a)1150.67 (n/a)1071.02 (n/a)1036.00 (n/a)1019.11 (n/a)60.48 (n/a)

gemm_896x1792x640_32x64x80_8cols_ccolmaj

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
1531781 — 2026-01-23 18:49:246.80 (+27.89%)6.43 (+24.70%)6.70 (+30.77%)5.33 (+5.74%)0.62 (+474.69%)1247.10 (-5.44%)1042.70 (-19.16%)992.90 (-23.53%)978.30 (-21.81%)114.59 (+329.89%)2100.72 (+27.89%)1987.76 (+24.70%)2069.85 (+30.77%)1647.93 (+5.74%)190.83 (+474.69%)
0f0da0b — 2026-01-14 16:31:005.32 (n/a)5.16 (n/a)5.12 (n/a)5.04 (n/a)0.11 (n/a)1318.80 (n/a)1289.76 (n/a)1298.50 (n/a)1251.20 (n/a)26.66 (n/a)1642.57 (n/a)1594.05 (n/a)1582.78 (n/a)1558.44 (n/a)33.21 (n/a)

input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
a4c2152 — 2026-02-09 21:03:120.08 (n/a)0.06 (n/a)0.06 (n/a)0.03 (n/a)0.01 (n/a)258.70 (n/a)153.13 (n/a)139.00 (n/a)106.30 (n/a)38.06 (n/a)

input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048-group_size_32

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
a4c2152 — 2026-02-09 21:03:120.08 (n/a)0.05 (n/a)0.04 (n/a)0.03 (n/a)0.02 (n/a)166.40 (n/a)117.12 (n/a)122.10 (n/a)66.30 (n/a)36.08 (n/a)

input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048-weighted_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
a4c2152 — 2026-02-09 21:03:120.07 (n/a)0.06 (n/a)0.07 (n/a)0.05 (n/a)0.01 (n/a)164.30 (n/a)130.28 (n/a)122.80 (n/a)113.00 (n/a)20.35 (n/a)

input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
a4c2152 — 2026-02-09 21:03:120.07 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)171.90 (n/a)135.14 (n/a)134.80 (n/a)114.00 (n/a)17.66 (n/a)

input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024-group_size_32

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
a4c2152 — 2026-02-09 21:03:120.04 (n/a)0.03 (n/a)0.03 (n/a)0.02 (n/a)0.01 (n/a)215.30 (n/a)164.42 (n/a)156.60 (n/a)130.40 (n/a)31.80 (n/a)

input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024-weighted_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
a4c2152 — 2026-02-09 21:03:120.06 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)209.80 (n/a)167.94 (n/a)179.30 (n/a)129.00 (n/a)36.07 (n/a)

input_length_2048-num_aie_columns_1-num_channels_2-tile_size_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
a4c2152 — 2026-02-09 21:03:120.11 (n/a)0.09 (n/a)0.09 (n/a)0.06 (n/a)0.01 (n/a)192.60 (n/a)137.55 (n/a)132.65 (n/a)115.20 (n/a)21.69 (n/a)

input_length_2048-num_aie_columns_1-num_channels_2-tile_size_2048-scalar_factor_3.0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
a4c2152 — 2026-02-09 21:03:120.09 (n/a)0.08 (n/a)0.08 (n/a)0.07 (n/a)0.01 (n/a)166.20 (n/a)151.88 (n/a)155.90 (n/a)133.30 (n/a)12.22 (n/a)

input_length_2048-num_aie_columns_1-num_channels_2-tile_size_2048-weighted_True

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
a4c2152 — 2026-02-09 21:03:120.10 (n/a)0.08 (n/a)0.08 (n/a)0.07 (n/a)0.01 (n/a)177.60 (n/a)154.48 (n/a)153.30 (n/a)123.00 (n/a)23.50 (n/a)

input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
a4c2152 — 2026-02-09 21:03:120.07 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)216.50 (n/a)153.53 (n/a)151.35 (n/a)117.10 (n/a)24.77 (n/a)

input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024-group_size_32

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
a4c2152 — 2026-02-09 21:03:120.05 (n/a)0.03 (n/a)0.03 (n/a)0.03 (n/a)0.01 (n/a)190.00 (n/a)163.36 (n/a)186.20 (n/a)108.00 (n/a)36.37 (n/a)

input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024-weighted_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
a4c2152 — 2026-02-09 21:03:120.07 (n/a)0.06 (n/a)0.05 (n/a)0.05 (n/a)0.01 (n/a)163.10 (n/a)147.60 (n/a)152.10 (n/a)117.10 (n/a)17.66 (n/a)

input_length_2048-num_aie_columns_2-num_channels_2-tile_size_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
a4c2152 — 2026-02-09 21:03:120.13 (n/a)0.09 (n/a)0.08 (n/a)0.06 (n/a)0.02 (n/a)196.20 (n/a)151.45 (n/a)145.80 (n/a)93.00 (n/a)33.58 (n/a)

input_length_2048-num_aie_columns_2-num_channels_2-tile_size_1024-scalar_factor_3.0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
a4c2152 — 2026-02-09 21:03:120.10 (n/a)0.09 (n/a)0.09 (n/a)0.07 (n/a)0.01 (n/a)171.40 (n/a)137.42 (n/a)133.80 (n/a)118.70 (n/a)21.62 (n/a)

input_length_2048-num_aie_columns_2-num_channels_2-tile_size_1024-weighted_True

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
a4c2152 — 2026-02-09 21:03:120.08 (n/a)0.07 (n/a)0.07 (n/a)0.05 (n/a)0.01 (n/a)192.60 (n/a)158.40 (n/a)157.40 (n/a)124.90 (n/a)23.99 (n/a)

input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
a4c2152 — 2026-02-09 21:03:120.07 (n/a)0.05 (n/a)0.06 (n/a)0.03 (n/a)0.01 (n/a)258.20 (n/a)163.41 (n/a)140.55 (n/a)117.50 (n/a)49.06 (n/a)

input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512-group_size_32

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
a4c2152 — 2026-02-09 21:03:120.04 (n/a)0.03 (n/a)0.03 (n/a)0.02 (n/a)0.01 (n/a)239.60 (n/a)189.70 (n/a)199.10 (n/a)124.20 (n/a)42.08 (n/a)

input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512-weighted_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
a4c2152 — 2026-02-09 21:03:120.07 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)209.30 (n/a)163.50 (n/a)158.30 (n/a)121.10 (n/a)34.85 (n/a)

input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
a4c2152 — 2026-02-09 21:03:120.08 (n/a)0.05 (n/a)0.05 (n/a)0.03 (n/a)0.01 (n/a)297.50 (n/a)164.10 (n/a)162.50 (n/a)106.30 (n/a)38.98 (n/a)

input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512-group_size_32

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
a4c2152 — 2026-02-09 21:03:120.05 (n/a)0.03 (n/a)0.03 (n/a)0.02 (n/a)0.01 (n/a)220.50 (n/a)166.36 (n/a)158.30 (n/a)101.60 (n/a)45.91 (n/a)

input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512-weighted_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
a4c2152 — 2026-02-09 21:03:120.07 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)186.40 (n/a)162.94 (n/a)181.20 (n/a)109.70 (n/a)32.97 (n/a)

input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
a4c2152 — 2026-02-09 21:03:120.06 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)223.60 (n/a)174.86 (n/a)166.80 (n/a)130.70 (n/a)28.05 (n/a)

input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256-group_size_32

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
a4c2152 — 2026-02-09 21:03:120.04 (n/a)0.03 (n/a)0.03 (n/a)0.03 (n/a)0.01 (n/a)203.40 (n/a)170.04 (n/a)186.40 (n/a)127.30 (n/a)36.80 (n/a)

input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256-weighted_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
a4c2152 — 2026-02-09 21:03:120.06 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)191.90 (n/a)167.56 (n/a)178.10 (n/a)137.10 (n/a)27.19 (n/a)

input_length_2048-num_aie_columns_4-num_channels_2-tile_size_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
a4c2152 — 2026-02-09 21:03:120.11 (n/a)0.07 (n/a)0.07 (n/a)0.03 (n/a)0.02 (n/a)386.50 (n/a)183.28 (n/a)179.55 (n/a)111.10 (n/a)77.31 (n/a)

input_length_2048-num_aie_columns_4-num_channels_2-tile_size_512-scalar_factor_3.0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
a4c2152 — 2026-02-09 21:03:120.09 (n/a)0.07 (n/a)0.06 (n/a)0.06 (n/a)0.01 (n/a)221.70 (n/a)187.88 (n/a)193.60 (n/a)137.80 (n/a)30.92 (n/a)

input_length_2048-num_aie_columns_4-num_channels_2-tile_size_512-weighted_True

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
a4c2152 — 2026-02-09 21:03:120.07 (n/a)0.06 (n/a)0.06 (n/a)0.06 (n/a)0.00 (n/a)166.40 (n/a)149.96 (n/a)147.10 (n/a)137.60 (n/a)10.57 (n/a)

input_length_2048-num_aie_columns_8-num_channels_1-tile_size_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
a4c2152 — 2026-02-09 21:03:120.07 (n/a)0.04 (n/a)0.05 (n/a)0.02 (n/a)0.01 (n/a)387.70 (n/a)191.59 (n/a)176.55 (n/a)122.50 (n/a)49.30 (n/a)

input_length_2048-num_aie_columns_8-num_channels_1-tile_size_256-group_size_32

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
a4c2152 — 2026-02-09 21:03:120.03 (n/a)0.03 (n/a)0.03 (n/a)0.02 (n/a)0.00 (n/a)217.40 (n/a)199.74 (n/a)208.60 (n/a)177.40 (n/a)18.18 (n/a)

input_length_2048-num_aie_columns_8-num_channels_1-tile_size_256-weighted_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
a4c2152 — 2026-02-09 21:03:120.05 (n/a)0.04 (n/a)0.04 (n/a)0.04 (n/a)0.01 (n/a)229.40 (n/a)186.36 (n/a)185.00 (n/a)151.90 (n/a)29.82 (n/a)

input_length_2048-num_aie_columns_8-num_channels_2-tile_size_128

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
a4c2152 — 2026-02-09 21:03:120.06 (n/a)0.04 (n/a)0.04 (n/a)0.03 (n/a)0.01 (n/a)237.20 (n/a)199.24 (n/a)201.10 (n/a)142.80 (n/a)29.66 (n/a)

input_length_2048-num_aie_columns_8-num_channels_2-tile_size_128-group_size_32

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
a4c2152 — 2026-02-09 21:03:120.03 (n/a)0.03 (n/a)0.03 (n/a)0.02 (n/a)0.00 (n/a)238.40 (n/a)205.60 (n/a)207.90 (n/a)168.10 (n/a)27.43 (n/a)

input_length_2048-num_aie_columns_8-num_channels_2-tile_size_128-weighted_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
a4c2152 — 2026-02-09 21:03:120.04 (n/a)0.04 (n/a)0.04 (n/a)0.03 (n/a)0.00 (n/a)234.10 (n/a)206.56 (n/a)198.80 (n/a)182.40 (n/a)20.78 (n/a)

input_length_2048-num_aie_columns_8-num_channels_2-tile_size_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
a4c2152 — 2026-02-09 21:03:120.10 (n/a)0.07 (n/a)0.07 (n/a)0.05 (n/a)0.01 (n/a)235.10 (n/a)176.85 (n/a)180.00 (n/a)123.70 (n/a)32.42 (n/a)

input_length_2048-num_aie_columns_8-num_channels_2-tile_size_256-scalar_factor_3.0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
a4c2152 — 2026-02-09 21:03:120.07 (n/a)0.05 (n/a)0.06 (n/a)0.03 (n/a)0.02 (n/a)421.00 (n/a)284.10 (n/a)214.10 (n/a)168.10 (n/a)123.98 (n/a)

input_length_2048-num_aie_columns_8-num_channels_2-tile_size_256-weighted_True

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
a4c2152 — 2026-02-09 21:03:120.06 (n/a)0.04 (n/a)0.04 (n/a)0.02 (n/a)0.01 (n/a)383.40 (n/a)240.00 (n/a)211.00 (n/a)155.90 (n/a)85.96 (n/a)

input_length_2048-num_cores_1-num_channels_1-bypass_False-tile_size_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
a4c2152 — 2026-02-09 21:03:120.08 (n/a)0.06 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)189.50 (n/a)151.18 (n/a)150.40 (n/a)106.30 (n/a)36.79 (n/a)

input_length_2048-num_cores_16-num_channels_2-bypass_False-tile_size_128

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
a4c2152 — 2026-02-09 21:03:120.06 (n/a)0.04 (n/a)0.04 (n/a)0.03 (n/a)0.01 (n/a)244.90 (n/a)202.82 (n/a)233.40 (n/a)142.10 (n/a)48.93 (n/a)

input_length_2048-num_cores_2-num_channels_1-bypass_False-tile_size_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
a4c2152 — 2026-02-09 21:03:120.07 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)156.90 (n/a)136.02 (n/a)133.30 (n/a)112.80 (n/a)16.84 (n/a)

input_length_2048-num_cores_2-num_channels_2-bypass_False-tile_size_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
a4c2152 — 2026-02-09 21:03:120.08 (n/a)0.06 (n/a)0.06 (n/a)0.04 (n/a)0.01 (n/a)193.60 (n/a)146.76 (n/a)147.30 (n/a)107.30 (n/a)31.01 (n/a)

input_length_2048-num_cores_4-num_channels_1-bypass_False-tile_size_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
a4c2152 — 2026-02-09 21:03:120.08 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)175.20 (n/a)141.64 (n/a)147.80 (n/a)109.20 (n/a)28.22 (n/a)

input_length_2048-num_cores_4-num_channels_2-bypass_False-tile_size_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
a4c2152 — 2026-02-09 21:03:120.08 (n/a)0.06 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)190.50 (n/a)151.28 (n/a)149.00 (n/a)106.30 (n/a)31.01 (n/a)

input_length_2048-num_cores_8-num_channels_1-bypass_False-tile_size_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
a4c2152 — 2026-02-09 21:03:120.05 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)232.50 (n/a)182.78 (n/a)170.90 (n/a)158.80 (n/a)31.09 (n/a)

input_length_2048-num_cores_8-num_channels_2-bypass_False-tile_size_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
a4c2152 — 2026-02-09 21:03:120.07 (n/a)0.06 (n/a)0.05 (n/a)0.05 (n/a)0.01 (n/a)165.50 (n/a)143.54 (n/a)158.80 (n/a)113.50 (n/a)24.61 (n/a)

input_length_32768-num_aie_columns_2-num_channels_2-tile_size_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
a4c2152 — 2026-02-09 21:03:121.06 (n/a)0.85 (n/a)0.83 (n/a)0.64 (n/a)0.19 (n/a)204.60 (n/a)161.52 (n/a)158.30 (n/a)124.00 (n/a)36.42 (n/a)

input_length_32768-num_aie_columns_2-num_channels_2-tile_size_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
a4c2152 — 2026-02-09 21:03:121.25 (n/a)0.89 (n/a)0.79 (n/a)0.68 (n/a)0.23 (n/a)192.90 (n/a)154.82 (n/a)165.90 (n/a)104.90 (n/a)35.60 (n/a)

input_length_32768-num_aie_columns_4-num_channels_4-tile_size_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
a4c2152 — 2026-02-09 21:03:120.77 (n/a)0.67 (n/a)0.66 (n/a)0.59 (n/a)0.07 (n/a)221.70 (n/a)198.06 (n/a)198.10 (n/a)169.50 (n/a)21.02 (n/a)

layer_norm_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.11 (+88.61%)0.09 (+61.18%)0.08 (+46.00%)0.07 (+61.81%)0.02 (+189.47%)119.70 (-38.20%)97.44 (-36.14%)100.50 (-31.54%)71.50 (-46.96%)22.67 (-5.36%)
0f0da0b — 2026-01-14 16:31:000.06 (n/a)0.05 (n/a)0.06 (n/a)0.04 (n/a)0.01 (n/a)193.70 (n/a)152.58 (n/a)146.80 (n/a)134.80 (n/a)23.96 (n/a)

layer_norm_1_cols_2_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.11 (+73.83%)0.09 (+50.07%)0.08 (+46.18%)0.06 (+7.28%)0.02 (+683.83%)138.10 (-6.82%)98.78 (-30.35%)97.10 (-31.57%)76.40 (-42.47%)24.88 (+313.86%)
0f0da0b — 2026-01-14 16:31:000.06 (n/a)0.06 (n/a)0.06 (n/a)0.06 (n/a)0.00 (n/a)148.20 (n/a)141.82 (n/a)141.90 (n/a)132.80 (n/a)6.01 (n/a)

layer_norm_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.12 (+58.00%)0.09 (+35.21%)0.09 (+33.88%)0.06 (+12.88%)0.02 (+142.40%)140.10 (-11.44%)98.38 (-23.32%)94.40 (-25.26%)68.10 (-36.71%)26.11 (+36.70%)
0f0da0b — 2026-01-14 16:31:000.08 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)158.20 (n/a)128.30 (n/a)126.30 (n/a)107.60 (n/a)19.10 (n/a)

layer_norm_2_cols_2_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.13 (+115.82%)0.09 (+79.97%)0.09 (+57.15%)0.06 (+84.94%)0.03 (+142.79%)130.40 (-45.94%)91.90 (-43.54%)93.50 (-36.39%)62.00 (-53.66%)25.59 (-42.10%)
0f0da0b — 2026-01-14 16:31:000.06 (n/a)0.05 (n/a)0.06 (n/a)0.03 (n/a)0.01 (n/a)241.20 (n/a)162.78 (n/a)147.00 (n/a)133.80 (n/a)44.20 (n/a)

layer_norm_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.11 (+63.95%)0.09 (+43.09%)0.11 (+55.71%)0.06 (-6.72%)0.02 (+806.48%)141.00 (+7.22%)91.54 (-25.97%)77.90 (-35.78%)73.00 (-39.07%)28.43 (+505.85%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.07 (n/a)0.07 (n/a)0.06 (n/a)0.00 (n/a)131.50 (n/a)123.66 (n/a)121.30 (n/a)119.80 (n/a)4.69 (n/a)

layer_norm_4_cols_2_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.11 (+76.69%)0.09 (+61.27%)0.09 (+70.26%)0.06 (+36.41%)0.02 (+195.74%)148.40 (-26.72%)103.14 (-34.74%)87.80 (-41.27%)73.70 (-43.44%)32.30 (+19.79%)
0f0da0b — 2026-01-14 16:31:000.06 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)202.50 (n/a)158.04 (n/a)149.50 (n/a)130.30 (n/a)26.96 (n/a)

layer_norm_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.12 (+75.09%)0.10 (+78.44%)0.11 (+94.76%)0.05 (+12.92%)0.03 (+213.89%)153.90 (-11.45%)90.36 (-39.81%)77.40 (-48.67%)67.30 (-42.87%)35.86 (+72.31%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.06 (n/a)0.05 (n/a)0.05 (n/a)0.01 (n/a)173.80 (n/a)150.12 (n/a)150.80 (n/a)117.80 (n/a)20.81 (n/a)

layer_norm_8_cols_2_channels_2048_tile_128

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.09 (+88.05%)0.08 (+72.40%)0.08 (+83.22%)0.06 (+32.78%)0.01 (+723.36%)140.30 (-24.69%)106.50 (-40.67%)97.60 (-45.44%)92.40 (-46.80%)19.84 (+234.33%)
0f0da0b — 2026-01-14 16:31:000.05 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.00 (n/a)186.30 (n/a)179.50 (n/a)178.90 (n/a)173.70 (n/a)5.93 (n/a)

matrix_vector_mul_128x128_32_1col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:490.37 (+17.13%)0.30 (+24.87%)0.30 (+38.03%)0.23 (+12.47%)0.06 (+35.23%)
0a6c11c — 2025-12-04 00:39:100.32 (n/a)0.24 (n/a)0.22 (n/a)0.20 (n/a)0.04 (n/a)

matrix_vector_mul_128x128_32_1col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:100.26 (-1.81%)0.24 (+4.06%)0.24 (+0.52%)0.21 (+16.97%)0.02 (-48.16%)0.26 (-1.81%)0.24 (+4.06%)0.24 (+0.52%)0.20 (+16.97%)0.02 (-48.16%)
a4b6ffe — 2026-01-05 19:24:050.26 (n/a)0.23 (n/a)0.24 (n/a)0.18 (n/a)0.04 (n/a)0.26 (n/a)0.23 (n/a)0.24 (n/a)0.17 (n/a)0.04 (n/a)

matrix_vector_mul_128x128_32tsi_128tso_1col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
1531781 — 2026-01-23 18:49:240.45 (+63.20%)0.35 (+43.38%)0.35 (+38.29%)0.22 (+12.34%)0.08 (+156.17%)0.44 (+63.20%)0.34 (+43.38%)0.34 (+38.29%)0.22 (+12.34%)0.08 (+156.17%)
0f0da0b — 2026-01-14 16:31:000.27 (n/a)0.24 (n/a)0.25 (n/a)0.20 (n/a)0.03 (n/a)0.27 (n/a)0.24 (n/a)0.25 (n/a)0.19 (n/a)0.03 (n/a)

matrix_vector_mul_2048x8192_1_1col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:4913.57 (+3.47%)13.15 (+1.09%)13.10 (+0.16%)12.52 (-1.58%)0.37 (+153.19%)
0a6c11c — 2025-12-04 00:39:1013.11 (n/a)13.01 (n/a)13.07 (n/a)12.72 (n/a)0.15 (n/a)

matrix_vector_mul_2048x8192_1_1col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:1013.43 (-0.07%)13.30 (+4.62%)13.27 (+1.29%)13.16 (+16.70%)0.12 (-86.69%)13.42 (-0.07%)13.29 (+4.62%)13.26 (+1.29%)13.16 (+16.70%)0.12 (-86.69%)
a4b6ffe — 2026-01-05 19:24:0513.44 (n/a)12.71 (n/a)13.10 (n/a)11.28 (n/a)0.87 (n/a)13.43 (n/a)12.70 (n/a)13.09 (n/a)11.27 (n/a)0.87 (n/a)

matrix_vector_mul_2048x8192_1_2col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:4925.06 (+2.47%)23.71 (+1.81%)24.13 (+0.50%)20.90 (+5.41%)1.48 (-15.60%)
0a6c11c — 2025-12-04 00:39:1024.46 (n/a)23.29 (n/a)24.01 (n/a)19.83 (n/a)1.75 (n/a)

matrix_vector_mul_2048x8192_1_2col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:1024.46 (-0.60%)21.07 (-8.03%)19.75 (-17.83%)18.08 (+1.18%)3.03 (+7.07%)24.44 (-0.60%)21.06 (-8.03%)19.74 (-17.83%)18.07 (+1.18%)3.03 (+7.07%)
a4b6ffe — 2026-01-05 19:24:0524.60 (n/a)22.91 (n/a)24.04 (n/a)17.87 (n/a)2.83 (n/a)24.59 (n/a)22.90 (n/a)24.02 (n/a)17.86 (n/a)2.83 (n/a)

matrix_vector_mul_2048x8192_1_4col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:4942.05 (+1.54%)40.50 (+12.60%)40.49 (+2.36%)38.98 (+97.53%)0.97 (-88.09%)
0a6c11c — 2025-12-04 00:39:1041.41 (n/a)35.97 (n/a)39.55 (n/a)19.73 (n/a)8.15 (n/a)

matrix_vector_mul_2048x8192_1_4col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:1042.58 (+1.51%)41.37 (+14.29%)41.25 (+4.89%)40.33 (+91.42%)0.92 (-89.18%)42.56 (+1.51%)41.34 (+14.29%)41.23 (+4.89%)40.30 (+91.42%)0.92 (-89.18%)
a4b6ffe — 2026-01-05 19:24:0541.95 (n/a)36.20 (n/a)39.33 (n/a)21.07 (n/a)8.54 (n/a)41.92 (n/a)36.17 (n/a)39.31 (n/a)21.05 (n/a)8.53 (n/a)

matrix_vector_mul_2048x8192_1_8col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:4946.62 (+3.11%)44.82 (+0.17%)44.78 (+0.17%)42.60 (-3.80%)1.60 (+367.72%)
0a6c11c — 2025-12-04 00:39:1045.21 (n/a)44.75 (n/a)44.70 (n/a)44.29 (n/a)0.34 (n/a)

matrix_vector_mul_2048x8192_1_8col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:1046.04 (+4.14%)45.06 (+3.47%)44.55 (+2.76%)44.30 (+2.84%)0.87 (+66.58%)46.01 (+4.14%)45.03 (+3.47%)44.52 (+2.76%)44.27 (+2.84%)0.87 (+66.58%)
a4b6ffe — 2026-01-05 19:24:0544.21 (n/a)43.55 (n/a)43.35 (n/a)43.07 (n/a)0.52 (n/a)44.18 (n/a)43.52 (n/a)43.32 (n/a)43.05 (n/a)0.52 (n/a)

matrix_vector_mul_2048x8192_1tsi_1024tso_2col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
1531781 — 2026-01-23 18:49:2424.80 (+0.16%)24.59 (-0.08%)24.55 (-0.22%)24.40 (+0.04%)0.18 (+11.75%)24.79 (+0.16%)24.58 (-0.08%)24.53 (-0.22%)24.38 (+0.04%)0.18 (+11.75%)
0f0da0b — 2026-01-14 16:31:0024.76 (n/a)24.61 (n/a)24.60 (n/a)24.39 (n/a)0.16 (n/a)24.75 (n/a)24.60 (n/a)24.58 (n/a)24.37 (n/a)0.16 (n/a)

matrix_vector_mul_2048x8192_1tsi_2048tso_1col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
1531781 — 2026-01-23 18:49:2413.54 (+0.77%)13.34 (+3.03%)13.45 (+0.75%)13.10 (+15.45%)0.21 (-77.02%)13.53 (+0.77%)13.33 (+3.03%)13.44 (+0.75%)13.09 (+15.45%)0.21 (-77.02%)
0f0da0b — 2026-01-14 16:31:0013.44 (n/a)12.95 (n/a)13.35 (n/a)11.35 (n/a)0.90 (n/a)13.43 (n/a)12.94 (n/a)13.34 (n/a)11.34 (n/a)0.90 (n/a)

matrix_vector_mul_2048x8192_1tsi_256tso_8col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
1531781 — 2026-01-23 18:49:2446.81 (+1.97%)45.91 (+2.21%)46.38 (+2.98%)44.81 (+2.36%)0.89 (-7.11%)46.79 (+1.97%)45.88 (+2.21%)46.35 (+2.98%)44.79 (+2.36%)0.89 (-7.11%)
0f0da0b — 2026-01-14 16:31:0045.91 (n/a)44.92 (n/a)45.04 (n/a)43.78 (n/a)0.96 (n/a)45.88 (n/a)44.89 (n/a)45.01 (n/a)43.75 (n/a)0.96 (n/a)

matrix_vector_mul_2048x8192_1tsi_512tso_4col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
1531781 — 2026-01-23 18:49:2441.49 (-0.92%)40.81 (+0.09%)41.13 (+0.38%)39.95 (+0.86%)0.65 (-24.40%)41.46 (-0.92%)40.79 (+0.09%)41.10 (+0.38%)39.93 (+0.86%)0.65 (-24.40%)
0f0da0b — 2026-01-14 16:31:0041.88 (n/a)40.78 (n/a)40.97 (n/a)39.61 (n/a)0.86 (n/a)41.85 (n/a)40.75 (n/a)40.95 (n/a)39.59 (n/a)0.86 (n/a)

matrix_vector_mul_8192x2048_4_1col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:4913.29 (+1.63%)13.19 (+6.08%)13.19 (+1.17%)13.09 (+26.86%)0.08 (-92.94%)
0a6c11c — 2025-12-04 00:39:1013.08 (n/a)12.44 (n/a)13.03 (n/a)10.32 (n/a)1.07 (n/a)

matrix_vector_mul_8192x2048_4_1col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:1013.68 (+3.42%)13.30 (+8.54%)13.38 (+7.25%)12.72 (+14.96%)0.35 (-66.09%)13.67 (+3.42%)13.29 (+8.54%)13.38 (+7.25%)12.71 (+14.96%)0.35 (-66.09%)
a4b6ffe — 2026-01-05 19:24:0513.23 (n/a)12.25 (n/a)12.48 (n/a)11.07 (n/a)1.04 (n/a)13.22 (n/a)12.24 (n/a)12.47 (n/a)11.06 (n/a)1.04 (n/a)

matrix_vector_mul_8192x2048_4_2col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:4925.58 (+4.24%)24.23 (+6.59%)24.03 (+0.29%)23.58 (+37.29%)0.70 (-74.97%)
0a6c11c — 2025-12-04 00:39:1024.54 (n/a)22.73 (n/a)23.96 (n/a)17.18 (n/a)2.79 (n/a)

matrix_vector_mul_8192x2048_4_2col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:1025.08 (+1.32%)24.68 (+13.42%)24.61 (+3.26%)24.47 (+48.32%)0.24 (-93.56%)25.07 (+1.32%)24.67 (+13.42%)24.59 (+3.26%)24.46 (+48.32%)0.24 (-93.56%)
a4b6ffe — 2026-01-05 19:24:0524.76 (n/a)21.76 (n/a)23.83 (n/a)16.50 (n/a)3.74 (n/a)24.74 (n/a)21.75 (n/a)23.82 (n/a)16.49 (n/a)3.73 (n/a)

matrix_vector_mul_8192x2048_4_4col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:4940.94 (-2.19%)39.74 (-1.10%)39.76 (-0.60%)38.17 (-2.14%)1.04 (+4.39%)
0a6c11c — 2025-12-04 00:39:1041.86 (n/a)40.18 (n/a)40.00 (n/a)39.00 (n/a)1.00 (n/a)

matrix_vector_mul_8192x2048_4_4col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:1040.74 (-1.09%)37.16 (-7.15%)40.25 (+1.47%)24.51 (-37.44%)7.07 (+736.13%)40.72 (-1.09%)37.13 (-7.15%)40.23 (+1.47%)24.50 (-37.44%)7.07 (+736.13%)
a4b6ffe — 2026-01-05 19:24:0541.19 (n/a)40.02 (n/a)39.67 (n/a)39.18 (n/a)0.85 (n/a)41.17 (n/a)39.99 (n/a)39.64 (n/a)39.16 (n/a)0.85 (n/a)

matrix_vector_mul_8192x2048_4_8col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:4946.71 (-1.04%)44.91 (-2.34%)44.71 (-2.98%)43.80 (-1.44%)0.96 (+6.93%)
0a6c11c — 2025-12-04 00:39:1047.20 (n/a)45.98 (n/a)46.08 (n/a)44.44 (n/a)0.90 (n/a)

matrix_vector_mul_8192x2048_4_8col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:1046.28 (+5.68%)44.89 (+14.59%)44.74 (+4.26%)44.20 (+62.80%)0.84 (-87.96%)46.25 (+5.68%)44.86 (+14.59%)44.71 (+4.26%)44.17 (+62.80%)0.84 (-87.96%)
a4b6ffe — 2026-01-05 19:24:0543.79 (n/a)39.17 (n/a)42.91 (n/a)27.15 (n/a)6.98 (n/a)43.77 (n/a)39.15 (n/a)42.88 (n/a)27.13 (n/a)6.98 (n/a)

matrix_vector_mul_8192x2048_4tsi_1024tso_1col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
1531781 — 2026-01-23 18:49:2413.46 (+0.53%)13.33 (+2.85%)13.32 (+1.17%)13.20 (+10.91%)0.11 (-81.73%)13.45 (+0.53%)13.32 (+2.85%)13.31 (+1.17%)13.19 (+10.91%)0.11 (-81.73%)
0f0da0b — 2026-01-14 16:31:0013.39 (n/a)12.96 (n/a)13.17 (n/a)11.90 (n/a)0.62 (n/a)13.38 (n/a)12.95 (n/a)13.16 (n/a)11.89 (n/a)0.62 (n/a)

matrix_vector_mul_8192x2048_4tsi_1024tso_2col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
1531781 — 2026-01-23 18:49:2425.27 (+0.55%)24.75 (-0.75%)24.61 (-1.28%)24.55 (-0.37%)0.30 (+52.22%)25.25 (+0.55%)24.74 (-0.75%)24.59 (-1.28%)24.53 (-0.37%)0.30 (+52.22%)
0f0da0b — 2026-01-14 16:31:0025.13 (n/a)24.94 (n/a)24.93 (n/a)24.64 (n/a)0.20 (n/a)25.11 (n/a)24.92 (n/a)24.91 (n/a)24.62 (n/a)0.20 (n/a)

matrix_vector_mul_8192x2048_4tsi_1024tso_4col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
1531781 — 2026-01-23 18:49:2440.86 (-4.37%)40.01 (-4.03%)40.43 (-3.32%)38.86 (-2.46%)0.89 (-24.84%)40.84 (-4.37%)39.98 (-4.03%)40.41 (-3.32%)38.84 (-2.46%)0.89 (-24.84%)
0f0da0b — 2026-01-14 16:31:0042.73 (n/a)41.69 (n/a)41.82 (n/a)39.85 (n/a)1.18 (n/a)42.70 (n/a)41.66 (n/a)41.80 (n/a)39.82 (n/a)1.18 (n/a)

matrix_vector_mul_8192x2048_4tsi_1024tso_8col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
1531781 — 2026-01-23 18:49:2446.08 (-1.82%)45.11 (-1.49%)45.49 (-0.25%)43.89 (-1.05%)0.97 (-6.46%)46.05 (-1.82%)45.08 (-1.49%)45.46 (-0.25%)43.87 (-1.05%)0.97 (-6.46%)
0f0da0b — 2026-01-14 16:31:0046.94 (n/a)45.79 (n/a)45.60 (n/a)44.36 (n/a)1.03 (n/a)46.91 (n/a)45.76 (n/a)45.58 (n/a)44.33 (n/a)1.03 (n/a)

mem_copy_16_cores_2_chans_2048_tile_128_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.07 (+22.03%)0.06 (+27.86%)0.06 (+35.49%)0.05 (+15.00%)0.01 (+49.69%)173.30 (-13.00%)137.94 (-21.32%)133.60 (-26.19%)123.40 (-18.06%)18.40 (+8.53%)
0a6c11c — 2025-12-04 00:39:100.05 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.00 (n/a)199.20 (n/a)175.32 (n/a)181.00 (n/a)150.60 (n/a)16.96 (n/a)

mem_copy_16_cores_2_chans_2048_tile_128_False0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.10 (+63.55%)0.08 (+85.69%)0.09 (+119.34%)0.04 (+23.15%)0.02 (+126.16%)195.30 (-18.79%)111.62 (-42.82%)86.60 (-54.42%)84.60 (-38.87%)47.65 (+9.68%)
0f0da0b — 2026-01-14 16:31:000.06 (n/a)0.04 (n/a)0.04 (n/a)0.03 (n/a)0.01 (n/a)240.50 (n/a)195.20 (n/a)190.00 (n/a)138.40 (n/a)43.45 (n/a)

mem_copy_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
003b49d — 2025-11-14 19:30:540.05 (-16.99%)0.05 (-12.47%)0.05 (-7.32%)0.04 (-11.17%)0.00 (-29.61%)197.00 (+12.57%)171.00 (+13.70%)164.00 (+7.89%)153.00 (+20.47%)17.49 (-4.32%)
0999a9a — 2025-11-13 19:24:080.06 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)175.00 (n/a)150.40 (n/a)152.00 (n/a)127.00 (n/a)18.28 (n/a)

mem_copy_1_cols_2_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
003b49d — 2025-11-14 19:30:540.06 (+5.56%)0.05 (-1.59%)0.05 (-1.39%)0.03 (-14.07%)0.01 (+25.03%)263.00 (+16.37%)166.00 (+4.53%)144.00 (+1.41%)126.00 (-5.26%)49.38 (+43.17%)
0999a9a — 2025-11-13 19:24:080.06 (n/a)0.05 (n/a)0.05 (n/a)0.03 (n/a)0.01 (n/a)226.00 (n/a)158.80 (n/a)142.00 (n/a)133.00 (n/a)34.49 (n/a)

mem_copy_1_cores_1_chans_2048_tile_2048_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.08 (+10.93%)0.07 (+7.79%)0.07 (+3.55%)0.05 (-1.73%)0.01 (+36.73%)150.20 (+1.76%)121.68 (-6.36%)125.60 (-3.38%)97.90 (-9.85%)18.56 (+21.98%)
0a6c11c — 2025-12-04 00:39:100.08 (n/a)0.06 (n/a)0.06 (n/a)0.06 (n/a)0.01 (n/a)147.60 (n/a)129.94 (n/a)130.00 (n/a)108.60 (n/a)15.21 (n/a)

mem_copy_1_cores_1_chans_2048_tile_2048_False0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.10 (+31.19%)0.09 (+26.03%)0.10 (+33.11%)0.07 (+25.01%)0.02 (+78.79%)118.10 (-20.04%)96.52 (-19.61%)84.40 (-24.91%)82.40 (-23.77%)18.29 (+9.09%)
0f0da0b — 2026-01-14 16:31:000.08 (n/a)0.07 (n/a)0.07 (n/a)0.06 (n/a)0.01 (n/a)147.70 (n/a)120.06 (n/a)112.40 (n/a)108.10 (n/a)16.76 (n/a)

mem_copy_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
003b49d — 2025-11-14 19:30:540.06 (-15.32%)0.05 (-17.09%)0.05 (-23.81%)0.04 (-15.54%)0.01 (-18.18%)193.00 (+18.40%)160.80 (+20.36%)168.00 (+31.25%)124.00 (+18.10%)22.92 (+11.39%)
0999a9a — 2025-11-13 19:24:080.07 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)163.00 (n/a)133.60 (n/a)128.00 (n/a)105.00 (n/a)20.58 (n/a)

mem_copy_2_cols_2_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
003b49d — 2025-11-14 19:30:540.06 (-23.02%)0.05 (-6.98%)0.05 (-2.13%)0.04 (-15.31%)0.01 (-33.29%)196.00 (+18.07%)147.00 (+6.52%)141.00 (+2.17%)126.00 (+29.90%)25.27 (+8.13%)
0999a9a — 2025-11-13 19:24:080.08 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)166.00 (n/a)138.00 (n/a)138.00 (n/a)97.00 (n/a)23.37 (n/a)

mem_copy_2_cores_1_chans_2048_tile_1024_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.10 (+31.37%)0.08 (+29.12%)0.08 (+29.06%)0.07 (+41.80%)0.01 (+0.23%)109.90 (-29.46%)98.30 (-23.38%)100.80 (-22.52%)79.40 (-23.87%)10.54 (-46.17%)
0a6c11c — 2025-12-04 00:39:100.08 (n/a)0.07 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)155.80 (n/a)128.30 (n/a)130.10 (n/a)104.30 (n/a)19.58 (n/a)

mem_copy_2_cores_1_chans_2048_tile_1024_False0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.12 (+103.55%)0.09 (+81.99%)0.09 (+69.86%)0.07 (+83.95%)0.02 (+187.42%)117.10 (-45.66%)92.20 (-43.53%)86.90 (-41.12%)70.40 (-50.84%)23.32 (-23.43%)
0f0da0b — 2026-01-14 16:31:000.06 (n/a)0.05 (n/a)0.06 (n/a)0.04 (n/a)0.01 (n/a)215.50 (n/a)163.26 (n/a)147.60 (n/a)143.20 (n/a)30.45 (n/a)

mem_copy_2_cores_2_chans_2048_tile_1024_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.10 (+25.02%)0.07 (+10.01%)0.07 (+7.53%)0.05 (-1.99%)0.01 (+77.08%)150.10 (+2.04%)114.60 (-7.22%)111.70 (-6.99%)83.90 (-20.02%)21.23 (+43.54%)
0a6c11c — 2025-12-04 00:39:100.08 (n/a)0.07 (n/a)0.07 (n/a)0.06 (n/a)0.01 (n/a)147.10 (n/a)123.52 (n/a)120.10 (n/a)104.90 (n/a)14.79 (n/a)

mem_copy_2_cores_2_chans_2048_tile_1024_False0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.10 (+35.25%)0.08 (+37.83%)0.09 (+48.91%)0.06 (+9.50%)0.02 (+125.91%)140.20 (-8.66%)101.66 (-25.39%)92.10 (-32.82%)82.90 (-26.11%)24.23 (+51.17%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)153.50 (n/a)136.26 (n/a)137.10 (n/a)112.20 (n/a)16.03 (n/a)

mem_copy_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
003b49d — 2025-11-14 19:30:540.07 (+11.97%)0.05 (+3.09%)0.05 (+2.10%)0.03 (-11.35%)0.01 (+41.65%)229.00 (+12.81%)156.00 (-0.51%)143.00 (-2.05%)117.00 (-10.69%)38.21 (+48.71%)
0999a9a — 2025-11-13 19:24:080.06 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)203.00 (n/a)156.80 (n/a)146.00 (n/a)131.00 (n/a)25.69 (n/a)

mem_copy_4_cols_2_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
003b49d — 2025-11-14 19:30:540.06 (+9.02%)0.05 (+5.59%)0.05 (+0.00%)0.04 (+3.63%)0.01 (+25.48%)193.00 (-3.50%)147.40 (-4.66%)147.00 (+0.00%)122.00 (-8.27%)25.24 (+7.48%)
0999a9a — 2025-11-13 19:24:080.06 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)200.00 (n/a)154.60 (n/a)147.00 (n/a)133.00 (n/a)23.48 (n/a)

mem_copy_4_cores_1_chans_2048_tile_512_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.09 (-3.98%)0.07 (+4.00%)0.07 (+7.04%)0.06 (+3.67%)0.01 (-15.76%)129.70 (-3.57%)113.30 (-4.50%)116.90 (-6.55%)90.40 (+4.15%)15.04 (-13.80%)
0a6c11c — 2025-12-04 00:39:100.09 (n/a)0.07 (n/a)0.07 (n/a)0.06 (n/a)0.01 (n/a)134.50 (n/a)118.64 (n/a)125.10 (n/a)86.80 (n/a)17.45 (n/a)

mem_copy_4_cores_1_chans_2048_tile_512_False0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.12 (+51.68%)0.10 (+55.58%)0.11 (+85.86%)0.06 (+8.92%)0.03 (+163.31%)128.30 (-8.23%)84.56 (-32.56%)71.60 (-46.17%)65.70 (-34.04%)26.42 (+57.86%)
0f0da0b — 2026-01-14 16:31:000.08 (n/a)0.07 (n/a)0.06 (n/a)0.06 (n/a)0.01 (n/a)139.80 (n/a)125.38 (n/a)133.00 (n/a)99.60 (n/a)16.73 (n/a)

mem_copy_4_cores_2_chans_2048_tile_512_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.08 (+9.57%)0.07 (+16.85%)0.07 (+19.58%)0.06 (+24.52%)0.01 (+1.05%)134.50 (-19.70%)116.64 (-14.92%)116.80 (-16.39%)98.00 (-8.75%)15.71 (-25.13%)
0a6c11c — 2025-12-04 00:39:100.08 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)167.50 (n/a)137.10 (n/a)139.70 (n/a)107.40 (n/a)20.99 (n/a)

mem_copy_4_cores_2_chans_2048_tile_512_False0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.10 (+21.52%)0.09 (+51.62%)0.10 (+79.69%)0.06 (+23.77%)0.02 (+25.48%)133.20 (-19.17%)92.98 (-33.88%)83.40 (-44.33%)82.10 (-17.74%)22.49 (-15.24%)
0f0da0b — 2026-01-14 16:31:000.08 (n/a)0.06 (n/a)0.05 (n/a)0.05 (n/a)0.01 (n/a)164.80 (n/a)140.62 (n/a)149.80 (n/a)99.80 (n/a)26.53 (n/a)

mem_copy_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
003b49d — 2025-11-14 19:30:540.05 (-13.48%)0.05 (-17.79%)0.05 (-15.33%)0.03 (-25.09%)0.01 (+6.67%)267.00 (+33.50%)176.80 (+24.51%)150.00 (+18.11%)141.00 (+15.57%)47.41 (+61.41%)
0999a9a — 2025-11-13 19:24:080.06 (n/a)0.06 (n/a)0.06 (n/a)0.04 (n/a)0.01 (n/a)200.00 (n/a)142.00 (n/a)127.00 (n/a)122.00 (n/a)29.37 (n/a)

mem_copy_8_cols_2_channels_2048_tile_128

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
003b49d — 2025-11-14 19:30:540.05 (+13.38%)0.05 (+23.85%)0.05 (+37.91%)0.05 (+30.72%)0.00 (-56.85%)166.00 (-23.50%)153.40 (-20.60%)153.00 (-27.49%)142.00 (-11.80%)7.63 (-70.60%)
0999a9a — 2025-11-13 19:24:080.05 (n/a)0.04 (n/a)0.04 (n/a)0.04 (n/a)0.01 (n/a)217.00 (n/a)193.20 (n/a)211.00 (n/a)161.00 (n/a)25.96 (n/a)

mem_copy_8_cores_1_chans_2048_tile_256_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.09 (+16.31%)0.08 (+15.34%)0.08 (+14.60%)0.07 (+19.30%)0.01 (-5.41%)117.30 (-16.15%)107.80 (-13.61%)108.50 (-12.78%)94.10 (-13.99%)8.29 (-32.01%)
0a6c11c — 2025-12-04 00:39:100.07 (n/a)0.07 (n/a)0.07 (n/a)0.06 (n/a)0.01 (n/a)139.90 (n/a)124.78 (n/a)124.40 (n/a)109.40 (n/a)12.20 (n/a)

mem_copy_8_cores_1_chans_2048_tile_256_False0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.09 (+46.69%)0.08 (+57.18%)0.09 (+64.02%)0.06 (+94.45%)0.01 (+6.58%)131.30 (-48.59%)99.78 (-38.77%)88.70 (-39.00%)86.90 (-31.84%)18.90 (-64.18%)
0f0da0b — 2026-01-14 16:31:000.06 (n/a)0.05 (n/a)0.06 (n/a)0.03 (n/a)0.01 (n/a)255.40 (n/a)162.96 (n/a)145.40 (n/a)127.50 (n/a)52.78 (n/a)

mem_copy_8_cores_2_chans_2048_tile_256_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.09 (+12.61%)0.07 (-2.24%)0.07 (-8.85%)0.06 (+5.13%)0.01 (+35.50%)136.50 (-4.88%)120.86 (+2.96%)125.10 (+9.64%)91.60 (-11.15%)15.72 (+10.46%)
0a6c11c — 2025-12-04 00:39:100.08 (n/a)0.07 (n/a)0.07 (n/a)0.06 (n/a)0.01 (n/a)143.50 (n/a)117.38 (n/a)114.10 (n/a)103.10 (n/a)14.23 (n/a)

mem_copy_8_cores_2_chans_2048_tile_256_False0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.10 (+63.97%)0.09 (+58.77%)0.09 (+53.52%)0.06 (+50.07%)0.02 (+70.34%)144.40 (-33.36%)97.70 (-36.49%)88.10 (-34.89%)80.00 (-39.02%)26.35 (-27.55%)
0f0da0b — 2026-01-14 16:31:000.06 (n/a)0.06 (n/a)0.06 (n/a)0.04 (n/a)0.01 (n/a)216.70 (n/a)153.84 (n/a)135.30 (n/a)131.20 (n/a)36.36 (n/a)

mha

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.18 (+0.16%)0.18 (+0.19%)0.18 (+0.12%)0.18 (+0.32%)0.00 (-52.24%)47253.90 (-0.32%)47213.70 (-0.19%)47210.90 (-0.12%)47167.70 (-0.16%)28.71 (-52.47%)
0a6c11c — 2025-12-04 00:39:100.18 (n/a)0.18 (n/a)0.18 (n/a)0.18 (n/a)0.00 (n/a)47405.00 (n/a)47301.72 (n/a)47265.90 (n/a)47241.80 (n/a)60.40 (n/a)

mha0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.18 (-0.08%)0.18 (+0.22%)0.18 (+0.32%)0.18 (+0.35%)0.00 (-42.58%)47367.50 (-0.34%)47232.68 (-0.22%)47232.00 (-0.32%)47073.00 (+0.08%)105.45 (-42.71%)
0f0da0b — 2026-01-14 16:31:000.18 (n/a)0.18 (n/a)0.18 (n/a)0.18 (n/a)0.00 (n/a)47531.20 (n/a)47335.18 (n/a)47384.30 (n/a)47037.20 (n/a)184.08 (n/a)

relu_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.11 (+69.81%)0.09 (+59.10%)0.10 (+75.49%)0.05 (+26.24%)0.03 (+206.46%)152.90 (-20.82%)101.62 (-32.81%)82.20 (-43.00%)73.00 (-41.13%)35.75 (+36.67%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.06 (n/a)0.06 (n/a)0.04 (n/a)0.01 (n/a)193.10 (n/a)151.24 (n/a)144.20 (n/a)124.00 (n/a)26.16 (n/a)

relu_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.10 (+60.88%)0.08 (+49.18%)0.09 (+55.13%)0.05 (+8.09%)0.02 (+316.89%)150.60 (-7.44%)101.26 (-30.06%)91.30 (-35.57%)83.00 (-37.87%)28.02 (+146.84%)
0f0da0b — 2026-01-14 16:31:000.06 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.00 (n/a)162.70 (n/a)144.78 (n/a)141.70 (n/a)133.60 (n/a)11.35 (n/a)

relu_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.10 (+36.44%)0.09 (+44.22%)0.09 (+49.00%)0.06 (+25.29%)0.02 (+65.69%)135.80 (-20.21%)96.20 (-29.75%)86.30 (-32.84%)83.10 (-26.72%)22.46 (-2.41%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)170.20 (n/a)136.94 (n/a)128.50 (n/a)113.40 (n/a)23.02 (n/a)

relu_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.11 (+58.81%)0.09 (+71.41%)0.10 (+74.46%)0.05 (+55.08%)0.02 (+79.07%)154.40 (-35.51%)93.32 (-40.67%)78.40 (-42.65%)75.80 (-37.04%)34.24 (-28.56%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.06 (n/a)0.06 (n/a)0.03 (n/a)0.01 (n/a)239.40 (n/a)157.28 (n/a)136.70 (n/a)120.40 (n/a)47.93 (n/a)

rms_norm_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.09 (+27.92%)0.08 (+35.45%)0.09 (+45.35%)0.06 (+20.87%)0.01 (+35.02%)131.50 (-17.30%)100.70 (-25.92%)92.40 (-31.20%)90.90 (-21.77%)17.35 (-10.74%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)159.00 (n/a)135.94 (n/a)134.30 (n/a)116.20 (n/a)19.43 (n/a)

rms_norm_1_cols_2_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.12 (+67.22%)0.10 (+57.83%)0.11 (+74.39%)0.06 (-3.10%)0.03 (+484.78%)145.20 (+3.20%)86.54 (-31.66%)72.10 (-42.64%)70.50 (-40.25%)32.80 (+266.62%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.06 (n/a)0.07 (n/a)0.06 (n/a)0.00 (n/a)140.70 (n/a)126.64 (n/a)125.70 (n/a)118.00 (n/a)8.95 (n/a)

rms_norm_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.09 (+26.55%)0.09 (+40.18%)0.09 (+47.01%)0.06 (+19.23%)0.01 (+54.24%)129.60 (-16.12%)98.22 (-28.12%)90.90 (-31.96%)89.80 (-20.95%)17.55 (+0.58%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)154.50 (n/a)136.64 (n/a)133.60 (n/a)113.60 (n/a)17.45 (n/a)

rms_norm_2_cols_2_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.13 (+67.61%)0.11 (+72.13%)0.11 (+66.55%)0.09 (+90.95%)0.02 (+33.97%)93.50 (-47.62%)74.32 (-42.73%)73.10 (-39.98%)62.40 (-40.34%)11.60 (-59.00%)
0f0da0b — 2026-01-14 16:31:000.08 (n/a)0.07 (n/a)0.07 (n/a)0.05 (n/a)0.01 (n/a)178.50 (n/a)129.78 (n/a)121.80 (n/a)104.60 (n/a)28.30 (n/a)

rms_norm_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.10 (+60.49%)0.09 (+76.01%)0.09 (+72.18%)0.06 (+64.97%)0.01 (+31.34%)130.70 (-39.41%)97.48 (-43.89%)90.20 (-41.92%)86.00 (-37.68%)18.66 (-50.69%)
0f0da0b — 2026-01-14 16:31:000.06 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)215.70 (n/a)173.74 (n/a)155.30 (n/a)138.00 (n/a)37.85 (n/a)

rms_norm_4_cols_2_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.11 (+60.28%)0.10 (+80.30%)0.11 (+105.91%)0.07 (+52.87%)0.02 (+90.94%)125.50 (-34.60%)83.90 (-43.66%)73.50 (-51.42%)72.90 (-37.64%)23.26 (-20.88%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.06 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)191.90 (n/a)148.92 (n/a)151.30 (n/a)116.90 (n/a)29.40 (n/a)

rms_norm_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.09 (+27.54%)0.08 (+37.27%)0.09 (+53.73%)0.05 (+11.36%)0.01 (+87.66%)154.30 (-10.24%)106.66 (-25.46%)94.80 (-34.98%)94.00 (-21.60%)26.64 (+33.21%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)171.90 (n/a)143.10 (n/a)145.80 (n/a)119.90 (n/a)20.00 (n/a)

rms_norm_8_cols_2_channels_2048_tile_128

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.09 (+38.93%)0.08 (+46.69%)0.09 (+73.48%)0.05 (+5.47%)0.02 (+140.19%)167.70 (-5.15%)109.36 (-28.83%)91.60 (-42.35%)88.90 (-28.02%)33.52 (+68.39%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.05 (n/a)0.05 (n/a)0.05 (n/a)0.01 (n/a)176.80 (n/a)153.66 (n/a)158.90 (n/a)123.50 (n/a)19.91 (n/a)

rope_1_cols_2_channels_4096_tile_4096_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1b2d230 — 2026-01-13 22:38:030.19 (-21.66%)0.17 (-8.53%)0.18 (+2.09%)0.14 (-6.71%)0.02 (-38.07%)174.10 (+7.20%)150.06 (+8.11%)139.60 (-2.04%)129.80 (+27.63%)20.36 (-10.65%)
d48746f — 2026-01-12 22:17:100.24 (n/a)0.18 (n/a)0.17 (n/a)0.15 (n/a)0.04 (n/a)162.40 (n/a)138.80 (n/a)142.50 (n/a)101.70 (n/a)22.79 (n/a)

rope_1c_32rows_512cols_32arows_0m

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:241.10 (+30.53%)0.99 (+42.49%)1.05 (+62.73%)0.73 (+21.76%)0.15 (+50.52%)133.80 (-17.86%)101.54 (-29.34%)93.30 (-38.58%)89.10 (-23.39%)18.60 (-3.62%)
0f0da0b — 2026-01-14 16:31:000.85 (n/a)0.69 (n/a)0.65 (n/a)0.60 (n/a)0.10 (n/a)162.90 (n/a)143.70 (n/a)151.90 (n/a)116.30 (n/a)19.30 (n/a)

rope_1c_32rows_512cols_8arows_0m

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.56 (-21.50%)0.52 (-8.95%)0.53 (-6.60%)0.48 (+7.05%)0.03 (-70.73%)153.70 (-6.57%)140.84 (+7.27%)140.10 (+7.03%)132.50 (+27.40%)8.40 (-65.00%)
0f0da0b — 2026-01-14 16:31:000.71 (n/a)0.58 (n/a)0.56 (n/a)0.45 (n/a)0.10 (n/a)164.50 (n/a)131.30 (n/a)130.90 (n/a)104.00 (n/a)24.00 (n/a)

rope_2_cols_2_channels_4096_tile_2048_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1b2d230 — 2026-01-13 22:38:030.21 (-0.56%)0.17 (-10.62%)0.16 (-13.58%)0.13 (-13.33%)0.03 (+17.05%)183.70 (+15.39%)151.76 (+12.77%)155.20 (+15.73%)117.30 (+0.60%)24.24 (+35.73%)
d48746f — 2026-01-12 22:17:100.21 (n/a)0.19 (n/a)0.18 (n/a)0.15 (n/a)0.02 (n/a)159.20 (n/a)134.58 (n/a)134.10 (n/a)116.60 (n/a)17.86 (n/a)

rope_2c_32rows_512cols_32arows_0m

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:241.25 (+43.66%)0.97 (+36.64%)0.91 (+36.32%)0.63 (+2.79%)0.25 (+150.75%)157.00 (-2.67%)108.10 (-23.42%)108.30 (-26.68%)78.70 (-30.42%)31.09 (+71.34%)
0f0da0b — 2026-01-14 16:31:000.87 (n/a)0.71 (n/a)0.67 (n/a)0.61 (n/a)0.10 (n/a)161.30 (n/a)141.16 (n/a)147.70 (n/a)113.10 (n/a)18.15 (n/a)

rope_2c_32rows_512cols_8arows_0m

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.63 (-10.72%)0.57 (+15.63%)0.57 (+20.61%)0.49 (+35.05%)0.05 (-60.20%)150.60 (-25.96%)130.58 (-17.08%)130.40 (-17.05%)116.80 (+12.09%)12.57 (-65.56%)
0f0da0b — 2026-01-14 16:31:000.71 (n/a)0.49 (n/a)0.47 (n/a)0.36 (n/a)0.13 (n/a)203.40 (n/a)157.48 (n/a)157.20 (n/a)104.20 (n/a)36.49 (n/a)

rope_4_cols_2_channels_4096_tile_1024_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1b2d230 — 2026-01-13 22:38:030.22 (+18.92%)0.17 (+6.24%)0.17 (+5.57%)0.14 (-0.50%)0.03 (+90.77%)171.10 (+0.47%)147.40 (-4.39%)148.10 (-5.31%)110.10 (-15.89%)23.59 (+60.67%)
d48746f — 2026-01-12 22:17:100.19 (n/a)0.16 (n/a)0.16 (n/a)0.14 (n/a)0.02 (n/a)170.30 (n/a)154.16 (n/a)156.40 (n/a)130.90 (n/a)14.69 (n/a)

rope_8_cols_2_channels_4096_tile_512_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1b2d230 — 2026-01-13 22:38:030.17 (-1.30%)0.15 (+6.63%)0.15 (-8.78%)0.14 (+39.68%)0.01 (-67.35%)174.90 (-28.41%)163.42 (-10.36%)167.40 (+9.63%)146.50 (+1.31%)11.19 (-76.17%)
d48746f — 2026-01-12 22:17:100.17 (n/a)0.14 (n/a)0.16 (n/a)0.10 (n/a)0.03 (n/a)244.30 (n/a)182.30 (n/a)152.70 (n/a)144.60 (n/a)46.98 (n/a)

rope_8c_32rows_512cols_32arows_0m

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:241.34 (+62.86%)0.89 (+47.92%)0.85 (+53.70%)0.60 (+49.43%)0.27 (+64.12%)163.50 (-33.07%)117.38 (-32.11%)115.50 (-34.93%)73.60 (-38.56%)32.08 (-33.78%)
0f0da0b — 2026-01-14 16:31:000.82 (n/a)0.60 (n/a)0.55 (n/a)0.40 (n/a)0.16 (n/a)244.30 (n/a)172.90 (n/a)177.50 (n/a)119.80 (n/a)48.44 (n/a)

rope_8c_32rows_512cols_8arows_0m

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.54 (-10.87%)0.52 (+6.74%)0.53 (+12.23%)0.48 (+19.79%)0.03 (-67.26%)152.90 (-16.49%)143.32 (-8.02%)139.70 (-10.91%)135.50 (+12.17%)7.46 (-68.88%)
0f0da0b — 2026-01-14 16:31:000.61 (n/a)0.48 (n/a)0.47 (n/a)0.40 (n/a)0.08 (n/a)183.10 (n/a)155.82 (n/a)156.80 (n/a)120.80 (n/a)23.99 (n/a)

rows_32-cols_512-angle_rows_32-aie_columns_1-method_type_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
a4c2152 — 2026-02-09 21:03:120.77 (n/a)0.72 (n/a)0.73 (n/a)0.65 (n/a)0.05 (n/a)151.90 (n/a)137.24 (n/a)134.20 (n/a)127.30 (n/a)9.16 (n/a)

rows_32-cols_512-angle_rows_32-aie_columns_2-method_type_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
a4c2152 — 2026-02-09 21:03:120.89 (n/a)0.75 (n/a)0.77 (n/a)0.60 (n/a)0.11 (n/a)164.50 (n/a)134.14 (n/a)128.30 (n/a)110.60 (n/a)20.19 (n/a)

rows_32-cols_512-angle_rows_32-aie_columns_8-method_type_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
a4c2152 — 2026-02-09 21:03:120.98 (n/a)0.67 (n/a)0.58 (n/a)0.48 (n/a)0.21 (n/a)206.90 (n/a)156.98 (n/a)170.70 (n/a)100.60 (n/a)44.52 (n/a)

rows_32-cols_512-angle_rows_8-aie_columns_1-method_type_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
a4c2152 — 2026-02-09 21:03:120.57 (n/a)0.46 (n/a)0.48 (n/a)0.28 (n/a)0.12 (n/a)264.90 (n/a)173.24 (n/a)152.80 (n/a)129.70 (n/a)56.52 (n/a)

rows_32-cols_512-angle_rows_8-aie_columns_2-method_type_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
a4c2152 — 2026-02-09 21:03:120.69 (n/a)0.59 (n/a)0.57 (n/a)0.53 (n/a)0.06 (n/a)138.50 (n/a)125.34 (n/a)129.20 (n/a)106.10 (n/a)12.09 (n/a)

rows_32-cols_512-angle_rows_8-aie_columns_8-method_type_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
a4c2152 — 2026-02-09 21:03:120.54 (n/a)0.45 (n/a)0.47 (n/a)0.38 (n/a)0.06 (n/a)191.90 (n/a)165.10 (n/a)158.50 (n/a)136.60 (n/a)23.18 (n/a)

seq_len_16384-dim_64-num_heads_1-num_pipelines_8

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
a4c2152 — 2026-02-09 21:03:120.18 (n/a)0.18 (n/a)0.18 (n/a)0.18 (n/a)0.00 (n/a)47629.10 (n/a)47464.16 (n/a)47509.90 (n/a)47089.30 (n/a)216.88 (n/a)

seq_len_256-embedding_dim_2048-hidden_dim_2048-prio_accuracy_False

No metrics available.

sigmoid_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.12 (+53.53%)0.10 (+54.49%)0.12 (+106.52%)0.05 (-0.16%)0.03 (+144.85%)153.20 (+0.13%)90.92 (-30.39%)68.60 (-51.55%)68.30 (-34.89%)36.80 (+58.03%)
0f0da0b — 2026-01-14 16:31:000.08 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)153.00 (n/a)130.62 (n/a)141.60 (n/a)104.90 (n/a)23.29 (n/a)

sigmoid_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.12 (+73.69%)0.10 (+60.60%)0.12 (+98.01%)0.06 (+2.86%)0.03 (+394.96%)140.80 (-2.83%)89.84 (-32.71%)68.90 (-49.52%)68.10 (-42.43%)32.04 (+163.16%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.06 (n/a)0.06 (n/a)0.06 (n/a)0.01 (n/a)144.90 (n/a)133.52 (n/a)136.50 (n/a)118.30 (n/a)12.18 (n/a)

sigmoid_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.10 (+30.15%)0.08 (+37.20%)0.09 (+57.41%)0.05 (+2.94%)0.02 (+110.00%)169.40 (-2.87%)106.58 (-23.47%)87.40 (-36.48%)84.00 (-23.15%)36.37 (+56.68%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)174.40 (n/a)139.26 (n/a)137.60 (n/a)109.30 (n/a)23.21 (n/a)

sigmoid_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.12 (+109.66%)0.10 (+84.35%)0.12 (+117.89%)0.05 (+3.89%)0.03 (+649.09%)157.30 (-3.73%)89.74 (-39.88%)68.70 (-54.11%)65.90 (-52.35%)38.85 (+251.55%)
0f0da0b — 2026-01-14 16:31:000.06 (n/a)0.06 (n/a)0.05 (n/a)0.05 (n/a)0.00 (n/a)163.40 (n/a)149.28 (n/a)149.70 (n/a)138.30 (n/a)11.05 (n/a)

silu_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.10 (+44.80%)0.08 (+33.22%)0.10 (+66.93%)0.06 (-1.08%)0.02 (+301.40%)143.30 (+1.06%)104.88 (-20.01%)82.20 (-40.13%)80.60 (-30.93%)32.24 (+177.47%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.06 (n/a)0.06 (n/a)0.06 (n/a)0.01 (n/a)141.80 (n/a)131.12 (n/a)137.30 (n/a)116.70 (n/a)11.62 (n/a)

silu_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.10 (+34.12%)0.08 (+47.27%)0.10 (+69.15%)0.06 (+39.46%)0.02 (+73.50%)138.90 (-28.29%)103.10 (-30.67%)83.20 (-40.91%)81.60 (-25.41%)28.30 (-8.88%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.06 (n/a)0.06 (n/a)0.04 (n/a)0.01 (n/a)193.70 (n/a)148.70 (n/a)140.80 (n/a)109.40 (n/a)31.06 (n/a)

silu_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.12 (+42.12%)0.11 (+61.07%)0.12 (+92.74%)0.06 (+8.38%)0.03 (+104.00%)137.40 (-7.72%)83.36 (-34.98%)70.00 (-48.07%)68.10 (-29.65%)30.27 (+32.68%)
0f0da0b — 2026-01-14 16:31:000.08 (n/a)0.07 (n/a)0.06 (n/a)0.06 (n/a)0.01 (n/a)148.90 (n/a)128.20 (n/a)134.80 (n/a)96.80 (n/a)22.82 (n/a)

silu_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.12 (+35.33%)0.09 (+59.58%)0.09 (+78.47%)0.07 (+60.20%)0.02 (-1.76%)109.80 (-37.58%)89.20 (-38.82%)87.90 (-43.98%)71.10 (-26.09%)14.50 (-53.17%)
0f0da0b — 2026-01-14 16:31:000.09 (n/a)0.06 (n/a)0.05 (n/a)0.05 (n/a)0.02 (n/a)175.90 (n/a)145.80 (n/a)156.90 (n/a)96.20 (n/a)30.97 (n/a)

softmax_1_cols_2_channels_4096_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.21 (+20.19%)0.16 (+13.84%)0.16 (+8.19%)0.12 (+26.61%)0.04 (+17.18%)135.50 (-20.99%)104.68 (-12.65%)102.70 (-7.56%)76.50 (-16.85%)22.55 (-26.21%)
0f0da0b — 2026-01-14 16:31:000.18 (n/a)0.14 (n/a)0.15 (n/a)0.10 (n/a)0.03 (n/a)171.50 (n/a)119.84 (n/a)111.10 (n/a)92.00 (n/a)30.56 (n/a)

softmax_2_cols_2_channels_4096_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.22 (+40.26%)0.18 (+43.33%)0.20 (+64.08%)0.13 (+24.34%)0.04 (+68.98%)127.80 (-19.62%)94.74 (-28.95%)83.40 (-39.08%)73.90 (-28.67%)24.46 (-5.31%)
0f0da0b — 2026-01-14 16:31:000.16 (n/a)0.13 (n/a)0.12 (n/a)0.10 (n/a)0.03 (n/a)159.00 (n/a)133.34 (n/a)136.90 (n/a)103.60 (n/a)25.83 (n/a)

softmax_2_cols_2_channels_4096_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.24 (+85.21%)0.17 (+69.97%)0.15 (+54.16%)0.11 (+37.21%)0.05 (+203.82%)148.10 (-27.12%)102.78 (-37.88%)106.70 (-35.10%)69.60 (-45.96%)31.81 (+15.98%)
0f0da0b — 2026-01-14 16:31:000.13 (n/a)0.10 (n/a)0.10 (n/a)0.08 (n/a)0.02 (n/a)203.20 (n/a)165.46 (n/a)164.40 (n/a)128.80 (n/a)27.43 (n/a)

swiglu

No metrics available.

swiglu_decode_1x2048x2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.00 (-15.46%)0.00 (-27.20%)0.00 (-28.17%)0.00 (-33.90%)0.00 (+1834.83%)5187.20 (+51.28%)4720.84 (+38.38%)4750.90 (+39.22%)4015.10 (+18.29%)390.23 (+3298.45%)
0a6c11c — 2025-12-04 00:39:100.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)3428.80 (n/a)3411.62 (n/a)3412.60 (n/a)3394.20 (n/a)11.48 (n/a)

swiglu_decode_1x2048x2048_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.00 (+28.14%)0.00 (+20.30%)0.00 (+10.00%)0.00 (+11.66%)0.00 (+124.73%)5111.40 (-10.44%)4324.88 (-14.91%)4736.10 (-9.09%)3410.70 (-21.96%)843.97 (+53.51%)
0f0da0b — 2026-01-14 16:31:000.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)5707.20 (n/a)5083.00 (n/a)5209.60 (n/a)4370.50 (n/a)549.78 (n/a)

tanh_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.10 (+42.65%)0.08 (+33.21%)0.07 (+27.20%)0.06 (+13.46%)0.02 (+130.64%)140.80 (-11.89%)109.14 (-23.00%)115.90 (-21.37%)85.60 (-29.89%)23.52 (+37.00%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)159.80 (n/a)141.74 (n/a)147.40 (n/a)122.10 (n/a)17.17 (n/a)

tanh_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.11 (+41.27%)0.09 (+40.30%)0.09 (+45.94%)0.08 (+28.08%)0.01 (+83.15%)103.30 (-21.98%)88.02 (-28.44%)87.80 (-31.46%)77.80 (-29.21%)9.64 (+2.40%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.07 (n/a)0.06 (n/a)0.06 (n/a)0.01 (n/a)132.40 (n/a)123.00 (n/a)128.10 (n/a)109.90 (n/a)9.41 (n/a)

tanh_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.10 (+48.49%)0.09 (+56.51%)0.09 (+68.30%)0.07 (+54.99%)0.01 (+44.32%)121.40 (-35.46%)93.40 (-36.24%)86.30 (-40.61%)84.60 (-32.70%)15.80 (-36.89%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.06 (n/a)0.06 (n/a)0.04 (n/a)0.01 (n/a)188.10 (n/a)146.48 (n/a)145.30 (n/a)125.70 (n/a)25.04 (n/a)

tanh_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.11 (+64.36%)0.09 (+41.14%)0.10 (+56.21%)0.05 (-7.33%)0.03 (+346.69%)163.80 (+7.91%)102.90 (-22.05%)80.40 (-35.94%)71.70 (-39.13%)39.99 (+185.25%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.06 (n/a)0.07 (n/a)0.05 (n/a)0.01 (n/a)151.80 (n/a)132.00 (n/a)125.50 (n/a)117.80 (n/a)14.02 (n/a)

transpose_2048_M_64_N_1_cols_1_channels_64_m_64_n_8_s

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:493.93 (-7.02%)3.68 (+2.48%)3.66 (+1.82%)3.45 (+15.62%)0.15 (-63.64%)151.80 (-13.50%)142.70 (-3.59%)143.40 (-1.78%)133.50 (+7.49%)5.90 (-66.22%)
0a6c11c — 2025-12-04 00:39:104.22 (n/a)3.59 (n/a)3.59 (n/a)2.99 (n/a)0.42 (n/a)175.50 (n/a)148.02 (n/a)146.00 (n/a)124.20 (n/a)17.46 (n/a)

transpose_2048_M_64_N_1_cols_1_channels_64_m_64_n_8_s0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:244.95 (+35.28%)4.34 (+33.32%)4.37 (+25.58%)3.74 (+35.73%)0.50 (+11.62%)140.30 (-26.35%)121.98 (-25.39%)119.90 (-20.33%)106.00 (-26.08%)14.15 (-39.67%)
0f0da0b — 2026-01-14 16:31:003.66 (n/a)3.26 (n/a)3.48 (n/a)2.75 (n/a)0.45 (n/a)190.50 (n/a)163.48 (n/a)150.50 (n/a)143.40 (n/a)23.46 (n/a)

transpose_2048_M_64_N_1_cols_2_channels_64_m_64_n_8_s

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:495.11 (+15.57%)3.85 (-3.32%)3.40 (-18.64%)3.29 (+7.09%)0.71 (+49.14%)159.60 (-6.61%)140.18 (+4.80%)154.20 (+22.97%)102.60 (-13.49%)22.70 (+19.77%)
0a6c11c — 2025-12-04 00:39:104.42 (n/a)3.99 (n/a)4.18 (n/a)3.07 (n/a)0.48 (n/a)170.90 (n/a)133.76 (n/a)125.40 (n/a)118.60 (n/a)18.95 (n/a)

transpose_2048_M_64_N_1_cols_2_channels_64_m_64_n_8_s0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:246.07 (+44.96%)4.99 (+45.77%)5.09 (+52.05%)4.09 (+35.80%)0.74 (+53.13%)128.10 (-26.34%)106.86 (-31.23%)103.00 (-34.27%)86.30 (-31.02%)15.67 (-22.94%)
0f0da0b — 2026-01-14 16:31:004.19 (n/a)3.42 (n/a)3.35 (n/a)3.01 (n/a)0.48 (n/a)173.90 (n/a)155.38 (n/a)156.70 (n/a)125.10 (n/a)20.34 (n/a)

weighted_rms_norm_1_cols_2_channels_2048_weights_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.11 (+41.27%)0.09 (+39.11%)0.09 (+47.90%)0.05 (+1.24%)0.02 (+132.19%)155.00 (-1.21%)99.92 (-24.84%)89.60 (-32.43%)76.50 (-29.17%)31.36 (+74.29%)
0f0da0b — 2026-01-14 16:31:000.08 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)156.90 (n/a)132.94 (n/a)132.60 (n/a)108.00 (n/a)17.99 (n/a)

weighted_rms_norm_2_cols_2_channels_2048_weights_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.10 (+48.28%)0.09 (+48.11%)0.09 (+60.52%)0.05 (+2.82%)0.02 (+134.54%)166.70 (-2.74%)102.72 (-28.96%)88.50 (-37.68%)83.10 (-32.55%)35.87 (+61.58%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)171.40 (n/a)144.60 (n/a)142.00 (n/a)123.20 (n/a)22.20 (n/a)

weighted_rms_norm_4_cols_2_channels_2048_weights_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.09 (+21.03%)0.09 (+31.88%)0.09 (+37.19%)0.07 (+25.91%)0.01 (+25.73%)119.00 (-20.61%)96.48 (-24.14%)89.90 (-27.09%)89.30 (-17.31%)12.81 (-17.86%)
0f0da0b — 2026-01-14 16:31:000.08 (n/a)0.07 (n/a)0.07 (n/a)0.05 (n/a)0.01 (n/a)149.90 (n/a)127.18 (n/a)123.30 (n/a)108.00 (n/a)15.59 (n/a)

weighted_rms_norm_8_cols_2_channels_2048_weights_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.11 (+67.16%)0.10 (+80.58%)0.10 (+80.44%)0.06 (+74.63%)0.02 (+72.16%)135.80 (-42.72%)89.92 (-44.57%)80.00 (-44.60%)74.30 (-40.23%)25.76 (-41.27%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.05 (n/a)0.06 (n/a)0.03 (n/a)0.01 (n/a)237.10 (n/a)162.22 (n/a)144.40 (n/a)124.30 (n/a)43.86 (n/a)

@github-actions
Copy link
Contributor

github-actions bot commented Feb 9, 2026

📊 Test Results for Test Example Applications

a4c2152 (2026_02_09_21_14_03)

IRONCLAD

Tested on 2026_02_09_21_14_03 at commit a4c2152.

Test Checks TTFT (mean)TPS (mean)
llama_3.2_1b_prompt_13_tokens_1 ❌ 0/1 n/a n/a
llama_3.2_1b_prompt_13_tokens_40 ❌ 0/1 n/a n/a
llama_3.2_1b_prompt_2048_tokens_1 ❌ 0/1 n/a n/a
llama_3.2_1b_prompt_2048_tokens_40 ❌ 0/1 n/a n/a
📈 Trends (vs main branch) for Test Example Applications

a4c2152 (2026_02_09_21_14_03)

IRONCLAD Trends

llama_3.2_1b

Commit/Date Num Tokens (max)Num Tokens (mean)Num Tokens (median)Num Tokens (min)Num Tokens (stddev)TPS (max)TPS (mean)TPS (median)TPS (min)TPS (stddev)TTFT (max)TTFT (mean)TTFT (median)TTFT (min)TTFT (stddev)Total (max)Total (mean)Total (median)Total (min)Total (stddev)
130b6ea — 2025-12-05 21:33:1240.00 (+0.00%)40.00 (+0.00%)40.00 (+0.00%)40.00 (+0.00%)0.00 (n/a)4.71 (-0.42%)4.64 (-0.09%)4.64 (+0.65%)4.55 (-0.22%)0.05 (-17.66%)4.41 (-0.34%)4.39 (-0.19%)4.38 (-0.33%)4.37 (-0.15%)0.01 (-25.90%)12.96 (-0.00%)12.80 (+0.07%)12.80 (-0.23%)12.67 (+0.44%)0.09 (-21.12%)
0a6c11c — 2025-12-03 23:35:1540.00 (n/a)40.00 (n/a)40.00 (n/a)40.00 (n/a)0.00 (n/a)4.73 (n/a)4.64 (n/a)4.61 (n/a)4.56 (n/a)0.06 (n/a)4.42 (n/a)4.40 (n/a)4.40 (n/a)4.37 (n/a)0.02 (n/a)12.96 (n/a)12.79 (n/a)12.83 (n/a)12.62 (n/a)0.12 (n/a)

llama_3.2_1b_prompt_13_tokens_1

Commit/Date Num_Tokens (max)Num_Tokens (mean)Num_Tokens (median)Num_Tokens (min)Num_Tokens (stddev)TPS (max)TPS (mean)TPS (median)TPS (min)TPS (stddev)TTFT (max)TTFT (mean)TTFT (median)TTFT (min)TTFT (stddev)
a4c2152 — 2026-02-09 21:13:32n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)
1531781 — 2026-01-23 18:40:341.00 (n/a)1.00 (n/a)1.00 (n/a)1.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.61 (n/a)0.61 (n/a)0.61 (n/a)0.61 (n/a)0.00 (n/a)

llama_3.2_1b_prompt_13_tokens_40

Commit/Date Num_Tokens (max)Num_Tokens (mean)Num_Tokens (median)Num_Tokens (min)Num_Tokens (stddev)TPS (max)TPS (mean)TPS (median)TPS (min)TPS (stddev)TTFT (max)TTFT (mean)TTFT (median)TTFT (min)TTFT (stddev)
a4c2152 — 2026-02-09 21:13:32n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)
1531781 — 2026-01-23 18:40:3440.00 (n/a)40.00 (n/a)40.00 (n/a)40.00 (n/a)0.00 (n/a)4.34 (n/a)4.34 (n/a)4.34 (n/a)4.34 (n/a)0.00 (n/a)0.61 (n/a)0.61 (n/a)0.61 (n/a)0.61 (n/a)0.00 (n/a)

llama_3.2_1b_prompt_2048_tokens_1

Commit/Date Num_Tokens (max)Num_Tokens (mean)Num_Tokens (median)Num_Tokens (min)Num_Tokens (stddev)TPS (max)TPS (mean)TPS (median)TPS (min)TPS (stddev)TTFT (max)TTFT (mean)TTFT (median)TTFT (min)TTFT (stddev)
a4c2152 — 2026-02-09 21:13:32n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)
1531781 — 2026-01-23 18:40:341.00 (n/a)1.00 (n/a)1.00 (n/a)1.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)2.70 (n/a)2.70 (n/a)2.70 (n/a)2.70 (n/a)0.00 (n/a)

llama_3.2_1b_prompt_2048_tokens_40

Commit/Date Num_Tokens (max)Num_Tokens (mean)Num_Tokens (median)Num_Tokens (min)Num_Tokens (stddev)TPS (max)TPS (mean)TPS (median)TPS (min)TPS (stddev)TTFT (max)TTFT (mean)TTFT (median)TTFT (min)TTFT (stddev)
a4c2152 — 2026-02-09 21:13:32n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)
1531781 — 2026-01-23 18:40:3440.00 (n/a)40.00 (n/a)40.00 (n/a)40.00 (n/a)0.00 (n/a)4.22 (n/a)4.22 (n/a)4.22 (n/a)4.22 (n/a)0.00 (n/a)2.70 (n/a)2.70 (n/a)2.70 (n/a)2.70 (n/a)0.00 (n/a)

@github-actions
Copy link
Contributor

github-actions bot commented Feb 10, 2026

📊 Test Results for Test Example Applications

3cb43db (2026_02_10_00_59_52)

IRONCLAD

Tested on 2026_02_10_00_59_52 at commit 3cb43db.

Test Checks TTFT (mean)TPS (mean)
llama_3.2_1b_prompt_13_tokens_1 ❌ 0/1 n/a n/a
llama_3.2_1b_prompt_13_tokens_40 ❌ 0/1 n/a n/a
llama_3.2_1b_prompt_2048_tokens_1 ❌ 0/1 n/a n/a
llama_3.2_1b_prompt_2048_tokens_40 ❌ 0/1 n/a n/a
📈 Trends (vs main branch) for Test Example Applications

3cb43db (2026_02_10_00_59_52)

IRONCLAD Trends

llama_3.2_1b

Commit/Date Num Tokens (max)Num Tokens (mean)Num Tokens (median)Num Tokens (min)Num Tokens (stddev)TPS (max)TPS (mean)TPS (median)TPS (min)TPS (stddev)TTFT (max)TTFT (mean)TTFT (median)TTFT (min)TTFT (stddev)Total (max)Total (mean)Total (median)Total (min)Total (stddev)
130b6ea — 2025-12-05 21:33:1240.00 (+0.00%)40.00 (+0.00%)40.00 (+0.00%)40.00 (+0.00%)0.00 (n/a)4.71 (-0.42%)4.64 (-0.09%)4.64 (+0.65%)4.55 (-0.22%)0.05 (-17.66%)4.41 (-0.34%)4.39 (-0.19%)4.38 (-0.33%)4.37 (-0.15%)0.01 (-25.90%)12.96 (-0.00%)12.80 (+0.07%)12.80 (-0.23%)12.67 (+0.44%)0.09 (-21.12%)
0a6c11c — 2025-12-03 23:35:1540.00 (n/a)40.00 (n/a)40.00 (n/a)40.00 (n/a)0.00 (n/a)4.73 (n/a)4.64 (n/a)4.61 (n/a)4.56 (n/a)0.06 (n/a)4.42 (n/a)4.40 (n/a)4.40 (n/a)4.37 (n/a)0.02 (n/a)12.96 (n/a)12.79 (n/a)12.83 (n/a)12.62 (n/a)0.12 (n/a)

llama_3.2_1b_prompt_13_tokens_1

Commit/Date Num_Tokens (max)Num_Tokens (mean)Num_Tokens (median)Num_Tokens (min)Num_Tokens (stddev)TPS (max)TPS (mean)TPS (median)TPS (min)TPS (stddev)TTFT (max)TTFT (mean)TTFT (median)TTFT (min)TTFT (stddev)
3cb43db — 2026-02-10 00:59:22n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)
1531781 — 2026-01-23 18:40:341.00 (n/a)1.00 (n/a)1.00 (n/a)1.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.61 (n/a)0.61 (n/a)0.61 (n/a)0.61 (n/a)0.00 (n/a)

llama_3.2_1b_prompt_13_tokens_40

Commit/Date Num_Tokens (max)Num_Tokens (mean)Num_Tokens (median)Num_Tokens (min)Num_Tokens (stddev)TPS (max)TPS (mean)TPS (median)TPS (min)TPS (stddev)TTFT (max)TTFT (mean)TTFT (median)TTFT (min)TTFT (stddev)
3cb43db — 2026-02-10 00:59:22n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)
1531781 — 2026-01-23 18:40:3440.00 (n/a)40.00 (n/a)40.00 (n/a)40.00 (n/a)0.00 (n/a)4.34 (n/a)4.34 (n/a)4.34 (n/a)4.34 (n/a)0.00 (n/a)0.61 (n/a)0.61 (n/a)0.61 (n/a)0.61 (n/a)0.00 (n/a)

llama_3.2_1b_prompt_2048_tokens_1

Commit/Date Num_Tokens (max)Num_Tokens (mean)Num_Tokens (median)Num_Tokens (min)Num_Tokens (stddev)TPS (max)TPS (mean)TPS (median)TPS (min)TPS (stddev)TTFT (max)TTFT (mean)TTFT (median)TTFT (min)TTFT (stddev)
3cb43db — 2026-02-10 00:59:22n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)
1531781 — 2026-01-23 18:40:341.00 (n/a)1.00 (n/a)1.00 (n/a)1.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)2.70 (n/a)2.70 (n/a)2.70 (n/a)2.70 (n/a)0.00 (n/a)

llama_3.2_1b_prompt_2048_tokens_40

Commit/Date Num_Tokens (max)Num_Tokens (mean)Num_Tokens (median)Num_Tokens (min)Num_Tokens (stddev)TPS (max)TPS (mean)TPS (median)TPS (min)TPS (stddev)TTFT (max)TTFT (mean)TTFT (median)TTFT (min)TTFT (stddev)
3cb43db — 2026-02-10 00:59:22n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)n/a (n/a)
1531781 — 2026-01-23 18:40:3440.00 (n/a)40.00 (n/a)40.00 (n/a)40.00 (n/a)0.00 (n/a)4.22 (n/a)4.22 (n/a)4.22 (n/a)4.22 (n/a)0.00 (n/a)2.70 (n/a)2.70 (n/a)2.70 (n/a)2.70 (n/a)0.00 (n/a)

@github-actions
Copy link
Contributor

github-actions bot commented Feb 10, 2026

📊 Test Results for Small Benchmark/Test Suite

3cb43db (2026_02_10_01_11_25)

IRONCLAD

Tested on 2026_02_10_01_11_25 at commit 3cb43db.

Test Checks Latency (mean)Bandwidth (mean)GFLOP/s (mean)
M_128-K_128-num_aie_columns_1-tile_size_input_32-tile_size_output_128 ✅ 5/5 n/a 0.11 n/a
M_1792-K_896-N_1152-num_aie_columns_8-b_col_maj_False-c_col_maj_True-m_64-k_32-n_48-trace_size_0-partition_N_1 ✅ 5/5 2497.24 3.81 n/a
M_192-K_384-N_64-num_aie_columns_4-b_col_maj_False-c_col_maj_False-m_48-k_96-n_16-trace_size_0-partition_N_1 ✅ 5/5 511.08 0.59 n/a
M_192-K_384-N_64-num_aie_columns_4-b_col_maj_True-c_col_maj_True-m_48-k_96-n_16-trace_size_0-partition_N_1 ✅ 5/5 728.64 0.45 n/a
M_2048-K_2048-N_2048-num_aie_columns_1-b_col_maj_False-c_col_maj_False-m_64-k_64-n_64-trace_size_0-partition_N_1 ✅ 5/5 49336.96 0.51 n/a
M_2048-K_2048-N_2048-num_aie_columns_2-b_col_maj_True-c_col_maj_False-m_64-k_64-n_64-trace_size_0-partition_N_1 ✅ 5/5 28497.02 0.88 n/a
M_2048-K_2048-N_2048-num_aie_columns_8-b_col_maj_True-c_col_maj_True-m_64-k_64-n_64-trace_size_0-partition_N_1 ✅ 5/5 8646.98 2.95 n/a
M_2048-K_8192-num_aie_columns_1-tile_size_input_1-tile_size_output_2048 ✅ 5/5 n/a 11.77 n/a
M_2048-K_8192-num_aie_columns_2-tile_size_input_1-tile_size_output_1024 ✅ 5/5 n/a 21.55 n/a
M_2048-K_8192-num_aie_columns_4-tile_size_input_1-tile_size_output_512 ✅ 5/5 n/a 31.76 n/a
M_2048-K_8192-num_aie_columns_8-tile_size_input_1-tile_size_output_256 ✅ 5/5 n/a 31.90 n/a
M_2048-N_64-aie_columns_1-channels_1-m_64-n_64-s_8 ✅ 5/5 368.32 1.56 n/a
M_2048-N_64-aie_columns_1-channels_2-m_64-n_64-s_8 ✅ 5/5 359.16 1.54 n/a
M_384-K_1536-N_1792-num_aie_columns_4-b_col_maj_True-c_col_maj_False-m_32-k_48-n_64-trace_size_0-partition_N_1 ✅ 5/5 2433.50 3.41 n/a
M_8192-K_2048-num_aie_columns_1-tile_size_input_4-tile_size_output_1024 ✅ 5/5 n/a 11.69 n/a
M_8192-K_2048-num_aie_columns_2-tile_size_input_4-tile_size_output_1024 ✅ 5/5 n/a 20.96 n/a
M_8192-K_2048-num_aie_columns_4-tile_size_input_4-tile_size_output_1024 ✅ 5/5 n/a 29.15 n/a
M_8192-K_2048-num_aie_columns_8-tile_size_input_4-tile_size_output_1024 ✅ 5/5 n/a 32.40 n/a
M_896-K_1792-N_640-num_aie_columns_8-b_col_maj_False-c_col_maj_True-m_32-k_64-n_80-trace_size_0-partition_N_1 ✅ 5/5 1337.38 5.09 n/a
embedding_dim_2048-hidden_dim_2048 ✅ 5/5 n/a n/a n/a
input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048 ✅ 30/30 285.18 0.03 n/a
input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048-group_size_32 ✅ 5/5 386.32 0.02 n/a
input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048-weighted_False ✅ 5/5 274.42 0.03 n/a
input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024 ✅ 10/10 264.56 0.03 n/a
input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024-group_size_32 ✅ 5/5 278.08 0.02 n/a
input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024-weighted_False ✅ 5/5 270.96 0.03 n/a
input_length_2048-num_aie_columns_1-num_channels_2-tile_size_2048 ✅ 10/10 322.22 0.04 n/a
input_length_2048-num_aie_columns_1-num_channels_2-tile_size_2048-scalar_factor_3.0 ✅ 5/5 271.02 0.05 n/a
input_length_2048-num_aie_columns_1-num_channels_2-tile_size_2048-weighted_True ✅ 5/5 269.12 0.05 n/a
input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024 ✅ 30/30 276.63 0.03 n/a
input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024-group_size_32 ✅ 5/5 291.12 0.02 n/a
input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024-weighted_False ✅ 5/5 283.90 0.03 n/a
input_length_2048-num_aie_columns_2-num_channels_2-tile_size_1024 ✅ 10/10 314.66 0.05 n/a
input_length_2048-num_aie_columns_2-num_channels_2-tile_size_1024-scalar_factor_3.0 ✅ 5/5 299.58 0.05 n/a
input_length_2048-num_aie_columns_2-num_channels_2-tile_size_1024-weighted_True ✅ 5/5 263.14 0.04 n/a
input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512 ✅ 10/10 318.76 0.03 n/a
input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512-group_size_32 ✅ 5/5 277.76 0.02 n/a
input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512-weighted_False ✅ 5/5 289.06 0.03 n/a
input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512 ✅ 30/30 275.53 0.03 n/a
input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512-group_size_32 ✅ 5/5 274.98 0.02 n/a
input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512-weighted_False ✅ 5/5 318.26 0.03 n/a
input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256 ✅ 10/10 307.56 0.03 n/a
input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256-group_size_32 ✅ 5/5 375.90 0.02 n/a
input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256-weighted_False ✅ 5/5 371.22 0.03 n/a
input_length_2048-num_aie_columns_4-num_channels_2-tile_size_512 ✅ 10/10 296.16 0.05 n/a
input_length_2048-num_aie_columns_4-num_channels_2-tile_size_512-scalar_factor_3.0 ✅ 5/5 273.22 0.05 n/a
input_length_2048-num_aie_columns_4-num_channels_2-tile_size_512-weighted_True ✅ 5/5 270.72 0.03 n/a
input_length_2048-num_aie_columns_8-num_channels_1-tile_size_256 ✅ 30/30 286.43 0.03 n/a
input_length_2048-num_aie_columns_8-num_channels_1-tile_size_256-group_size_32 ✅ 5/5 278.38 0.02 n/a
input_length_2048-num_aie_columns_8-num_channels_1-tile_size_256-weighted_False ✅ 5/5 290.88 0.03 n/a
input_length_2048-num_aie_columns_8-num_channels_2-tile_size_128 ✅ 10/10 363.25 0.03 n/a
input_length_2048-num_aie_columns_8-num_channels_2-tile_size_128-group_size_32 ✅ 5/5 350.14 0.02 n/a
input_length_2048-num_aie_columns_8-num_channels_2-tile_size_128-weighted_False ✅ 5/5 400.64 0.03 n/a
input_length_2048-num_aie_columns_8-num_channels_2-tile_size_256 ✅ 10/10 330.94 0.04 n/a
input_length_2048-num_aie_columns_8-num_channels_2-tile_size_256-scalar_factor_3.0 ✅ 5/5 325.12 0.04 n/a
input_length_2048-num_aie_columns_8-num_channels_2-tile_size_256-weighted_True ✅ 5/5 305.26 0.03 n/a
input_length_2048-num_cores_1-num_channels_1-bypass_False-tile_size_2048 ✅ 5/5 275.36 0.03 n/a
input_length_2048-num_cores_16-num_channels_2-bypass_False-tile_size_128 ✅ 5/5 384.00 0.03 n/a
input_length_2048-num_cores_2-num_channels_1-bypass_False-tile_size_1024 ✅ 5/5 257.58 0.03 n/a
input_length_2048-num_cores_2-num_channels_2-bypass_False-tile_size_1024 ✅ 5/5 248.68 0.03 n/a
input_length_2048-num_cores_4-num_channels_1-bypass_False-tile_size_512 ✅ 5/5 265.18 0.03 n/a
input_length_2048-num_cores_4-num_channels_2-bypass_False-tile_size_512 ✅ 5/5 283.70 0.03 n/a
input_length_2048-num_cores_8-num_channels_1-bypass_False-tile_size_256 ✅ 5/5 322.62 0.03 n/a
input_length_2048-num_cores_8-num_channels_2-bypass_False-tile_size_256 ✅ 5/5 293.02 0.03 n/a
input_length_32768-num_aie_columns_2-num_channels_2-tile_size_1024 ✅ 5/5 329.88 0.43 n/a
input_length_32768-num_aie_columns_2-num_channels_2-tile_size_512 ✅ 5/5 279.52 0.48 n/a
input_length_32768-num_aie_columns_4-num_channels_4-tile_size_2048 ✅ 5/5 344.52 0.44 n/a
rows_32-cols_512-angle_rows_32-aie_columns_1-method_type_0 ✅ 5/5 370.30 0.27 n/a
rows_32-cols_512-angle_rows_32-aie_columns_2-method_type_0 ✅ 5/5 333.86 0.30 n/a
rows_32-cols_512-angle_rows_32-aie_columns_8-method_type_0 ✅ 5/5 393.58 0.26 n/a
rows_32-cols_512-angle_rows_8-aie_columns_1-method_type_0 ✅ 5/5 318.20 0.23 n/a
rows_32-cols_512-angle_rows_8-aie_columns_2-method_type_0 ✅ 5/5 321.02 0.24 n/a
rows_32-cols_512-angle_rows_8-aie_columns_8-method_type_0 ✅ 5/5 460.86 0.17 n/a
seq_len_16384-dim_64-num_heads_1-num_pipelines_8 ✅ 5/5 48327.64 0.17 n/a
seq_len_256-embedding_dim_2048-hidden_dim_2048-prio_accuracy_False ❌ 0/5 n/a n/a n/a
📈 Trends (vs main branch) for Small Benchmark/Test Suite

3cb43db (2026_02_10_01_11_25)

IRONCLAD Trends

M_128-K_128-num_aie_columns_1-tile_size_input_32-tile_size_output_128

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
3cb43db — 2026-02-10 01:03:390.14 (n/a)0.11 (n/a)0.13 (n/a)0.04 (n/a)0.04 (n/a)0.13 (n/a)0.11 (n/a)0.13 (n/a)0.04 (n/a)0.04 (n/a)

M_1792-K_896-N_1152-num_aie_columns_8-b_col_maj_False-c_col_maj_True-m_64-k_32-n_48-trace_size_0-partition_N_1

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
3cb43db — 2026-02-10 01:03:394.57 (n/a)3.81 (n/a)3.74 (n/a)3.29 (n/a)0.47 (n/a)2859.60 (n/a)2497.24 (n/a)2512.70 (n/a)2059.90 (n/a)289.31 (n/a)1795.87 (n/a)1498.41 (n/a)1472.28 (n/a)1293.67 (n/a)184.89 (n/a)

M_192-K_384-N_64-num_aie_columns_4-b_col_maj_False-c_col_maj_False-m_48-k_96-n_16-trace_size_0-partition_N_1

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
3cb43db — 2026-02-10 01:03:390.84 (n/a)0.59 (n/a)0.75 (n/a)0.22 (n/a)0.31 (n/a)998.30 (n/a)511.08 (n/a)294.70 (n/a)262.00 (n/a)339.10 (n/a)36.02 (n/a)25.26 (n/a)32.02 (n/a)9.45 (n/a)13.06 (n/a)

M_192-K_384-N_64-num_aie_columns_4-b_col_maj_True-c_col_maj_True-m_48-k_96-n_16-trace_size_0-partition_N_1

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
3cb43db — 2026-02-10 01:03:390.81 (n/a)0.45 (n/a)0.36 (n/a)0.13 (n/a)0.27 (n/a)1681.30 (n/a)728.64 (n/a)618.90 (n/a)272.30 (n/a)562.72 (n/a)34.66 (n/a)19.04 (n/a)15.25 (n/a)5.61 (n/a)11.49 (n/a)

M_2048-K_2048-N_2048-num_aie_columns_1-b_col_maj_False-c_col_maj_False-m_64-k_64-n_64-trace_size_0-partition_N_1

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
3cb43db — 2026-02-10 01:03:390.51 (n/a)0.51 (n/a)0.51 (n/a)0.51 (n/a)0.00 (n/a)49477.80 (n/a)49336.96 (n/a)49325.70 (n/a)49248.00 (n/a)85.85 (n/a)348.84 (n/a)348.22 (n/a)348.29 (n/a)347.22 (n/a)0.61 (n/a)

M_2048-K_2048-N_2048-num_aie_columns_2-b_col_maj_True-c_col_maj_False-m_64-k_64-n_64-trace_size_0-partition_N_1

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
3cb43db — 2026-02-10 01:03:390.89 (n/a)0.88 (n/a)0.88 (n/a)0.88 (n/a)0.00 (n/a)28676.70 (n/a)28497.02 (n/a)28478.10 (n/a)28303.90 (n/a)135.75 (n/a)606.98 (n/a)602.88 (n/a)603.26 (n/a)599.09 (n/a)2.87 (n/a)

M_2048-K_2048-N_2048-num_aie_columns_8-b_col_maj_True-c_col_maj_True-m_64-k_64-n_64-trace_size_0-partition_N_1

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
3cb43db — 2026-02-10 01:03:393.31 (n/a)2.95 (n/a)3.05 (n/a)2.32 (n/a)0.37 (n/a)10853.30 (n/a)8646.98 (n/a)8256.90 (n/a)7607.00 (n/a)1265.76 (n/a)2258.43 (n/a)2016.51 (n/a)2080.66 (n/a)1582.92 (n/a)254.53 (n/a)

M_2048-K_8192-num_aie_columns_1-tile_size_input_1-tile_size_output_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
3cb43db — 2026-02-10 01:03:3912.68 (n/a)11.77 (n/a)12.45 (n/a)10.49 (n/a)1.11 (n/a)12.68 (n/a)11.77 (n/a)12.44 (n/a)10.48 (n/a)1.11 (n/a)

M_2048-K_8192-num_aie_columns_2-tile_size_input_1-tile_size_output_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
3cb43db — 2026-02-10 01:03:3922.58 (n/a)21.55 (n/a)22.16 (n/a)18.77 (n/a)1.58 (n/a)22.56 (n/a)21.54 (n/a)22.15 (n/a)18.75 (n/a)1.58 (n/a)

M_2048-K_8192-num_aie_columns_4-tile_size_input_1-tile_size_output_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
3cb43db — 2026-02-10 01:03:3935.07 (n/a)31.76 (n/a)32.87 (n/a)25.80 (n/a)3.74 (n/a)35.04 (n/a)31.74 (n/a)32.85 (n/a)25.78 (n/a)3.73 (n/a)

M_2048-K_8192-num_aie_columns_8-tile_size_input_1-tile_size_output_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
3cb43db — 2026-02-10 01:03:3938.17 (n/a)31.90 (n/a)35.62 (n/a)14.03 (n/a)10.07 (n/a)38.15 (n/a)31.88 (n/a)35.59 (n/a)14.02 (n/a)10.06 (n/a)

M_2048-N_64-aie_columns_1-channels_1-m_64-n_64-s_8

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3cb43db — 2026-02-10 01:03:391.95 (n/a)1.56 (n/a)1.57 (n/a)0.84 (n/a)0.44 (n/a)624.40 (n/a)368.32 (n/a)334.30 (n/a)268.50 (n/a)146.64 (n/a)

M_2048-N_64-aie_columns_1-channels_2-m_64-n_64-s_8

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3cb43db — 2026-02-10 01:03:392.02 (n/a)1.54 (n/a)1.61 (n/a)1.06 (n/a)0.39 (n/a)495.70 (n/a)359.16 (n/a)325.20 (n/a)258.90 (n/a)95.57 (n/a)

M_384-K_1536-N_1792-num_aie_columns_4-b_col_maj_True-c_col_maj_False-m_32-k_48-n_64-trace_size_0-partition_N_1

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
3cb43db — 2026-02-10 01:03:394.02 (n/a)3.41 (n/a)3.55 (n/a)2.42 (n/a)0.60 (n/a)3324.30 (n/a)2433.50 (n/a)2269.00 (n/a)2005.10 (n/a)515.33 (n/a)1054.28 (n/a)894.98 (n/a)931.64 (n/a)635.91 (n/a)156.60 (n/a)

M_8192-K_2048-num_aie_columns_1-tile_size_input_4-tile_size_output_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
3cb43db — 2026-02-10 01:03:3912.60 (n/a)11.69 (n/a)11.50 (n/a)10.69 (n/a)0.85 (n/a)12.59 (n/a)11.68 (n/a)11.49 (n/a)10.69 (n/a)0.85 (n/a)

M_8192-K_2048-num_aie_columns_2-tile_size_input_4-tile_size_output_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
3cb43db — 2026-02-10 01:03:3922.00 (n/a)20.96 (n/a)21.66 (n/a)17.82 (n/a)1.76 (n/a)21.98 (n/a)20.95 (n/a)21.65 (n/a)17.81 (n/a)1.76 (n/a)

M_8192-K_2048-num_aie_columns_4-tile_size_input_4-tile_size_output_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
3cb43db — 2026-02-10 01:03:3933.54 (n/a)29.15 (n/a)30.62 (n/a)21.50 (n/a)4.97 (n/a)33.52 (n/a)29.13 (n/a)30.60 (n/a)21.49 (n/a)4.96 (n/a)

M_8192-K_2048-num_aie_columns_8-tile_size_input_4-tile_size_output_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
3cb43db — 2026-02-10 01:03:3937.02 (n/a)32.40 (n/a)35.64 (n/a)25.47 (n/a)5.49 (n/a)37.00 (n/a)32.38 (n/a)35.62 (n/a)25.45 (n/a)5.49 (n/a)

M_896-K_1792-N_640-num_aie_columns_8-b_col_maj_False-c_col_maj_True-m_32-k_64-n_80-trace_size_0-partition_N_1

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
3cb43db — 2026-02-10 01:03:395.93 (n/a)5.09 (n/a)5.29 (n/a)4.10 (n/a)0.84 (n/a)1622.20 (n/a)1337.38 (n/a)1258.10 (n/a)1121.80 (n/a)230.70 (n/a)1832.13 (n/a)1572.63 (n/a)1633.54 (n/a)1266.93 (n/a)260.77 (n/a)

axpy_1_cols_2_channels_2048_tile_2048_3.0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.13 (-1.96%)0.11 (+2.52%)0.10 (+1.86%)0.08 (-1.91%)0.02 (-1.72%)162.40 (+1.95%)120.96 (-2.36%)117.00 (-1.85%)94.40 (+2.05%)23.11 (+4.09%)
0a6c11c — 2025-12-04 00:39:100.13 (n/a)0.10 (n/a)0.10 (n/a)0.08 (n/a)0.02 (n/a)159.30 (n/a)123.88 (n/a)119.20 (n/a)92.50 (n/a)22.21 (n/a)

axpy_1_cols_2_channels_2048_tile_2048_3.0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.16 (+80.10%)0.13 (+54.74%)0.11 (+34.32%)0.10 (+39.85%)0.03 (+432.10%)118.80 (-28.48%)100.30 (-33.22%)110.40 (-25.56%)78.00 (-44.48%)20.49 (+105.98%)
0f0da0b — 2026-01-14 16:31:000.09 (n/a)0.08 (n/a)0.08 (n/a)0.07 (n/a)0.01 (n/a)166.10 (n/a)150.20 (n/a)148.30 (n/a)140.50 (n/a)9.95 (n/a)

axpy_2_cols_2_channels_2048_tile_1024_3.0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.11 (-26.77%)0.09 (-9.09%)0.09 (-10.81%)0.08 (+8.25%)0.01 (-55.42%)152.00 (-7.60%)135.46 (+6.28%)142.80 (+12.18%)115.00 (+36.58%)15.03 (-41.36%)
0a6c11c — 2025-12-04 00:39:100.15 (n/a)0.10 (n/a)0.10 (n/a)0.07 (n/a)0.02 (n/a)164.50 (n/a)127.46 (n/a)127.30 (n/a)84.20 (n/a)25.64 (n/a)

axpy_2_cols_2_channels_2048_tile_1024_3.0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.14 (+54.25%)0.12 (+45.65%)0.14 (+61.31%)0.08 (-2.79%)0.03 (+557.87%)156.90 (+2.89%)104.46 (-28.09%)90.30 (-37.98%)87.90 (-35.22%)29.52 (+351.33%)
0f0da0b — 2026-01-14 16:31:000.09 (n/a)0.08 (n/a)0.08 (n/a)0.08 (n/a)0.00 (n/a)152.50 (n/a)145.26 (n/a)145.60 (n/a)135.70 (n/a)6.54 (n/a)

axpy_4_cols_2_channels_2048_tile_512_3.0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.12 (-10.21%)0.10 (-8.41%)0.10 (-6.20%)0.07 (+3.52%)0.02 (-23.48%)166.70 (-3.36%)128.78 (+7.33%)121.80 (+6.65%)103.90 (+11.36%)22.95 (-18.82%)
0a6c11c — 2025-12-04 00:39:100.13 (n/a)0.11 (n/a)0.11 (n/a)0.07 (n/a)0.02 (n/a)172.50 (n/a)119.98 (n/a)114.20 (n/a)93.30 (n/a)28.27 (n/a)

axpy_4_cols_2_channels_2048_tile_512_3.0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.16 (+65.19%)0.14 (+57.55%)0.15 (+63.48%)0.09 (+24.56%)0.03 (+220.28%)134.10 (-19.70%)92.44 (-34.51%)83.40 (-38.81%)78.60 (-39.49%)23.38 (+58.01%)
0f0da0b — 2026-01-14 16:31:000.09 (n/a)0.09 (n/a)0.09 (n/a)0.07 (n/a)0.01 (n/a)167.00 (n/a)141.16 (n/a)136.30 (n/a)129.90 (n/a)14.80 (n/a)

axpy_8_cols_2_channels_2048_tile_256_3.0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.10 (-16.19%)0.08 (-10.40%)0.09 (-2.09%)0.07 (-17.38%)0.01 (-10.61%)184.40 (+21.00%)153.18 (+11.99%)144.10 (+2.20%)127.10 (+19.34%)21.26 (+34.76%)
0a6c11c — 2025-12-04 00:39:100.12 (n/a)0.09 (n/a)0.09 (n/a)0.08 (n/a)0.01 (n/a)152.40 (n/a)136.78 (n/a)141.00 (n/a)106.50 (n/a)15.78 (n/a)

axpy_8_cols_2_channels_2048_tile_256_3.0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.17 (+67.03%)0.14 (+66.86%)0.15 (+90.72%)0.08 (+20.53%)0.03 (+162.97%)146.80 (-17.06%)92.24 (-37.11%)80.00 (-47.54%)72.30 (-40.15%)31.07 (+36.92%)
0f0da0b — 2026-01-14 16:31:000.10 (n/a)0.09 (n/a)0.08 (n/a)0.07 (n/a)0.01 (n/a)177.00 (n/a)146.68 (n/a)152.50 (n/a)120.80 (n/a)22.69 (n/a)

dequant_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.05 (-18.83%)0.04 (-5.21%)0.04 (-2.69%)0.03 (+37.10%)0.00 (-53.98%)160.40 (-27.06%)141.12 (-0.91%)136.40 (+2.79%)114.30 (+23.17%)17.37 (-59.06%)
0a6c11c — 2025-12-04 00:39:100.06 (n/a)0.04 (n/a)0.04 (n/a)0.02 (n/a)0.01 (n/a)219.90 (n/a)142.42 (n/a)132.70 (n/a)92.80 (n/a)42.43 (n/a)

dequant_1_cols_1_channels_2048_tile_2048_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.06 (+13.11%)0.05 (+1.55%)0.04 (-0.75%)0.04 (+7.05%)0.01 (+25.85%)125.50 (-6.55%)115.08 (-1.12%)121.60 (+0.75%)88.10 (-11.63%)15.40 (+4.09%)
0f0da0b — 2026-01-14 16:31:000.05 (n/a)0.05 (n/a)0.04 (n/a)0.04 (n/a)0.01 (n/a)134.30 (n/a)116.38 (n/a)120.70 (n/a)99.70 (n/a)14.79 (n/a)

dequant_1_cols_2_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.06 (-13.28%)0.05 (-8.93%)0.06 (-13.59%)0.04 (+4.63%)0.01 (-39.10%)133.60 (-4.43%)106.08 (+5.85%)95.10 (+15.69%)87.90 (+15.35%)17.90 (-31.97%)
0a6c11c — 2025-12-04 00:39:100.07 (n/a)0.06 (n/a)0.06 (n/a)0.04 (n/a)0.01 (n/a)139.80 (n/a)100.22 (n/a)82.20 (n/a)76.20 (n/a)26.31 (n/a)

dequant_1_cols_2_channels_2048_tile_1024_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.07 (+24.45%)0.05 (+9.74%)0.05 (-2.94%)0.03 (+16.88%)0.01 (+14.24%)163.70 (-14.47%)114.66 (-9.57%)111.30 (+3.06%)76.50 (-19.64%)31.63 (-21.04%)
0f0da0b — 2026-01-14 16:31:000.06 (n/a)0.04 (n/a)0.05 (n/a)0.03 (n/a)0.01 (n/a)191.40 (n/a)126.80 (n/a)108.00 (n/a)95.20 (n/a)40.06 (n/a)

dequant_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.07 (+5.66%)0.05 (+1.21%)0.05 (+0.61%)0.04 (-15.02%)0.01 (+58.42%)143.40 (+17.64%)101.02 (+1.49%)97.60 (-0.61%)78.90 (-5.40%)22.56 (+78.52%)
0a6c11c — 2025-12-04 00:39:100.06 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)121.90 (n/a)99.54 (n/a)98.20 (n/a)83.40 (n/a)12.64 (n/a)

dequant_2_cols_1_channels_2048_tile_1024_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.05 (+2.33%)0.05 (+22.41%)0.05 (+32.19%)0.04 (+7.46%)0.01 (-3.28%)147.80 (-6.93%)113.14 (-18.51%)106.50 (-24.31%)101.40 (-2.31%)19.61 (-8.79%)
0f0da0b — 2026-01-14 16:31:000.05 (n/a)0.04 (n/a)0.04 (n/a)0.03 (n/a)0.01 (n/a)158.80 (n/a)138.84 (n/a)140.70 (n/a)103.80 (n/a)21.51 (n/a)

dequant_2_cols_2_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.05 (+17.06%)0.04 (+8.98%)0.05 (+21.37%)0.03 (-11.65%)0.01 (+301.89%)159.70 (+13.18%)126.32 (-5.68%)109.00 (-17.67%)106.80 (-14.56%)22.48 (+280.15%)
0a6c11c — 2025-12-04 00:39:100.04 (n/a)0.04 (n/a)0.04 (n/a)0.04 (n/a)0.00 (n/a)141.10 (n/a)133.92 (n/a)132.40 (n/a)125.00 (n/a)5.91 (n/a)

dequant_2_cols_2_channels_2048_tile_512_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.05 (+14.83%)0.04 (+10.04%)0.05 (+21.96%)0.03 (-12.01%)0.01 (+143.96%)158.30 (+13.64%)121.38 (-6.49%)106.20 (-17.99%)98.20 (-12.94%)26.13 (+141.52%)
0f0da0b — 2026-01-14 16:31:000.05 (n/a)0.04 (n/a)0.04 (n/a)0.04 (n/a)0.00 (n/a)139.30 (n/a)129.80 (n/a)129.50 (n/a)112.80 (n/a)10.82 (n/a)

dequant_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.06 (+20.24%)0.05 (-3.19%)0.05 (-6.73%)0.03 (-20.69%)0.01 (+194.26%)156.10 (+26.09%)120.86 (+7.15%)114.90 (+7.18%)88.30 (-16.85%)24.40 (+214.82%)
0a6c11c — 2025-12-04 00:39:100.05 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.00 (n/a)123.80 (n/a)112.80 (n/a)107.20 (n/a)106.20 (n/a)7.75 (n/a)

dequant_4_cols_1_channels_2048_tile_512_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.06 (+28.85%)0.05 (+16.79%)0.05 (+23.60%)0.03 (-12.66%)0.01 (+157.83%)164.80 (+14.52%)113.62 (-10.20%)103.20 (-19.12%)84.80 (-22.42%)32.55 (+130.11%)
0f0da0b — 2026-01-14 16:31:000.05 (n/a)0.04 (n/a)0.04 (n/a)0.04 (n/a)0.00 (n/a)143.90 (n/a)126.52 (n/a)127.60 (n/a)109.30 (n/a)14.15 (n/a)

dequant_4_cols_2_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.05 (-6.09%)0.04 (-8.04%)0.04 (-12.46%)0.03 (-0.41%)0.01 (-20.19%)168.10 (+0.42%)135.74 (+7.44%)132.20 (+14.26%)107.40 (+6.44%)22.91 (-12.67%)
0a6c11c — 2025-12-04 00:39:100.05 (n/a)0.04 (n/a)0.05 (n/a)0.03 (n/a)0.01 (n/a)167.40 (n/a)126.34 (n/a)115.70 (n/a)100.90 (n/a)26.23 (n/a)

dequant_4_cols_2_channels_2048_tile_256_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.07 (+42.90%)0.05 (+60.94%)0.06 (+84.55%)0.03 (+50.15%)0.01 (+27.03%)188.40 (-33.40%)112.72 (-39.08%)93.80 (-45.84%)79.70 (-29.96%)43.58 (-36.67%)
0f0da0b — 2026-01-14 16:31:000.05 (n/a)0.03 (n/a)0.03 (n/a)0.02 (n/a)0.01 (n/a)282.90 (n/a)185.04 (n/a)173.20 (n/a)113.80 (n/a)68.82 (n/a)

dequant_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.04 (-13.65%)0.04 (-13.67%)0.04 (-18.56%)0.03 (-13.52%)0.00 (-23.88%)170.40 (+15.68%)140.12 (+15.33%)135.80 (+22.78%)119.60 (+15.78%)17.94 (+1.87%)
0a6c11c — 2025-12-04 00:39:100.05 (n/a)0.04 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)147.30 (n/a)121.50 (n/a)110.60 (n/a)103.30 (n/a)17.61 (n/a)

dequant_8_cols_1_channels_2048_tile_256_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.04 (-5.92%)0.04 (-11.07%)0.04 (-12.74%)0.03 (-9.17%)0.00 (-4.85%)166.90 (+10.09%)138.66 (+12.51%)135.90 (+14.59%)117.60 (+6.33%)17.94 (+10.10%)
0f0da0b — 2026-01-14 16:31:000.05 (n/a)0.04 (n/a)0.04 (n/a)0.03 (n/a)0.00 (n/a)151.60 (n/a)123.24 (n/a)118.60 (n/a)110.60 (n/a)16.30 (n/a)

dequant_8_cols_2_channels_2048_tile_128

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.05 (+17.21%)0.03 (-1.70%)0.03 (-3.04%)0.03 (-16.20%)0.01 (+95.59%)198.20 (+19.33%)160.00 (+4.92%)166.20 (+3.17%)106.50 (-14.66%)29.86 (+87.19%)
0a6c11c — 2025-12-04 00:39:100.04 (n/a)0.03 (n/a)0.03 (n/a)0.03 (n/a)0.00 (n/a)166.10 (n/a)152.50 (n/a)161.10 (n/a)124.80 (n/a)15.95 (n/a)

dequant_8_cols_2_channels_2048_tile_128_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.05 (+10.58%)0.04 (+11.15%)0.04 (+32.75%)0.02 (-15.30%)0.01 (+23.92%)229.90 (+18.08%)145.60 (-7.34%)131.00 (-24.67%)106.20 (-9.54%)49.03 (+42.68%)
0f0da0b — 2026-01-14 16:31:000.04 (n/a)0.03 (n/a)0.03 (n/a)0.03 (n/a)0.01 (n/a)194.70 (n/a)157.14 (n/a)173.90 (n/a)117.40 (n/a)34.37 (n/a)

eltwise_add_1_cols_2_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.17 (+59.76%)0.12 (+36.17%)0.10 (+5.69%)0.08 (+17.86%)0.04 (+129.37%)155.80 (-15.19%)109.78 (-22.78%)120.50 (-5.34%)73.80 (-37.40%)34.38 (+16.64%)
0f0da0b — 2026-01-14 16:31:000.10 (n/a)0.09 (n/a)0.10 (n/a)0.07 (n/a)0.02 (n/a)183.70 (n/a)142.16 (n/a)127.30 (n/a)117.90 (n/a)29.48 (n/a)

eltwise_add_2_cols_2_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.15 (+24.10%)0.13 (+23.91%)0.13 (+30.32%)0.08 (+16.80%)0.03 (+31.09%)146.70 (-14.36%)101.20 (-18.68%)91.70 (-23.26%)81.40 (-19.41%)26.00 (-6.71%)
0f0da0b — 2026-01-14 16:31:000.12 (n/a)0.10 (n/a)0.10 (n/a)0.07 (n/a)0.02 (n/a)171.30 (n/a)124.44 (n/a)119.50 (n/a)101.00 (n/a)27.87 (n/a)

eltwise_add_4_cols_2_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.17 (+52.81%)0.15 (+77.83%)0.15 (+81.17%)0.11 (+106.70%)0.02 (+14.06%)116.00 (-51.65%)86.46 (-45.77%)80.90 (-44.82%)72.70 (-34.56%)17.22 (-64.39%)
0f0da0b — 2026-01-14 16:31:000.11 (n/a)0.08 (n/a)0.08 (n/a)0.05 (n/a)0.02 (n/a)239.90 (n/a)159.44 (n/a)146.60 (n/a)111.10 (n/a)48.37 (n/a)

eltwise_add_8_cols_2_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.16 (+58.53%)0.14 (+73.02%)0.14 (+70.95%)0.11 (+188.94%)0.02 (-23.24%)109.50 (-65.38%)90.00 (-48.14%)86.00 (-41.50%)75.30 (-36.93%)13.44 (-83.58%)
0f0da0b — 2026-01-14 16:31:000.10 (n/a)0.08 (n/a)0.08 (n/a)0.04 (n/a)0.03 (n/a)316.30 (n/a)173.56 (n/a)147.00 (n/a)119.40 (n/a)81.81 (n/a)

eltwise_mul_1_cols_2_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.14 (+42.19%)0.13 (+64.04%)0.13 (+74.51%)0.10 (+75.35%)0.01 (+0.96%)118.10 (-42.97%)97.88 (-39.87%)92.60 (-42.70%)90.40 (-29.70%)11.55 (-59.47%)
0f0da0b — 2026-01-14 16:31:000.10 (n/a)0.08 (n/a)0.08 (n/a)0.06 (n/a)0.01 (n/a)207.10 (n/a)162.78 (n/a)161.60 (n/a)128.60 (n/a)28.50 (n/a)

eltwise_mul_2_cols_2_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.13 (+44.68%)0.12 (+43.52%)0.13 (+47.71%)0.09 (+35.56%)0.02 (+80.76%)130.00 (-26.26%)102.10 (-29.85%)93.30 (-32.34%)91.10 (-30.88%)16.53 (-8.94%)
0f0da0b — 2026-01-14 16:31:000.09 (n/a)0.09 (n/a)0.09 (n/a)0.07 (n/a)0.01 (n/a)176.30 (n/a)145.54 (n/a)137.90 (n/a)131.80 (n/a)18.15 (n/a)

eltwise_mul_4_cols_2_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.17 (+63.62%)0.15 (+73.87%)0.15 (+72.82%)0.11 (+104.66%)0.02 (+10.66%)114.00 (-51.16%)86.18 (-44.46%)81.50 (-42.12%)74.20 (-38.88%)15.85 (-65.75%)
0f0da0b — 2026-01-14 16:31:000.10 (n/a)0.08 (n/a)0.09 (n/a)0.05 (n/a)0.02 (n/a)233.40 (n/a)155.18 (n/a)140.80 (n/a)121.40 (n/a)46.29 (n/a)

eltwise_mul_8_cols_2_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.13 (+46.12%)0.11 (+59.06%)0.12 (+66.26%)0.08 (+48.55%)0.02 (+55.56%)154.80 (-32.70%)112.28 (-36.89%)104.60 (-39.85%)95.50 (-31.59%)24.14 (-26.88%)
0f0da0b — 2026-01-14 16:31:000.09 (n/a)0.07 (n/a)0.07 (n/a)0.05 (n/a)0.01 (n/a)230.00 (n/a)177.92 (n/a)173.90 (n/a)139.60 (n/a)33.01 (n/a)

embedding_dim_2048-hidden_dim_2048

No metrics available.

gelu_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.11 (+53.19%)0.10 (+56.21%)0.11 (+82.01%)0.08 (+32.44%)0.02 (+220.41%)101.60 (-24.52%)83.80 (-34.61%)72.40 (-45.11%)72.20 (-34.72%)15.75 (+58.78%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.06 (n/a)0.06 (n/a)0.06 (n/a)0.01 (n/a)134.60 (n/a)128.16 (n/a)131.90 (n/a)110.60 (n/a)9.92 (n/a)

gelu_1_cols_2_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.12 (+63.09%)0.11 (+76.43%)0.12 (+100.14%)0.07 (+62.57%)0.02 (+62.52%)110.00 (-38.48%)80.26 (-43.34%)69.10 (-50.00%)66.40 (-38.69%)18.54 (-39.34%)
0f0da0b — 2026-01-14 16:31:000.08 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)178.80 (n/a)141.66 (n/a)138.20 (n/a)108.30 (n/a)30.55 (n/a)

gelu_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.12 (+51.06%)0.10 (+80.20%)0.12 (+120.49%)0.06 (+32.94%)0.03 (+101.39%)137.80 (-24.74%)85.02 (-42.29%)68.20 (-54.65%)67.80 (-33.79%)30.24 (+4.82%)
0f0da0b — 2026-01-14 16:31:000.08 (n/a)0.06 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)183.10 (n/a)147.32 (n/a)150.40 (n/a)102.40 (n/a)28.85 (n/a)

gelu_2_cols_2_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.12 (+49.86%)0.11 (+81.45%)0.12 (+78.91%)0.08 (+119.08%)0.02 (+8.63%)107.80 (-54.38%)76.80 (-47.80%)69.50 (-44.13%)67.10 (-33.23%)17.40 (-67.51%)
0f0da0b — 2026-01-14 16:31:000.08 (n/a)0.06 (n/a)0.07 (n/a)0.03 (n/a)0.02 (n/a)236.30 (n/a)147.12 (n/a)124.40 (n/a)100.50 (n/a)53.56 (n/a)

gelu_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.13 (+73.67%)0.11 (+92.11%)0.12 (+121.10%)0.05 (+47.52%)0.03 (+98.64%)153.10 (-32.20%)85.56 (-45.77%)69.30 (-54.77%)63.00 (-42.47%)37.94 (-17.18%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.06 (n/a)0.05 (n/a)0.04 (n/a)0.02 (n/a)225.80 (n/a)157.78 (n/a)153.20 (n/a)109.50 (n/a)45.81 (n/a)

gelu_4_cols_2_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.12 (+87.02%)0.11 (+117.65%)0.12 (+144.58%)0.07 (+103.45%)0.02 (+63.92%)110.50 (-50.82%)79.80 (-54.61%)70.50 (-59.08%)70.00 (-46.56%)17.48 (-56.79%)
0f0da0b — 2026-01-14 16:31:000.06 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)224.70 (n/a)175.80 (n/a)172.30 (n/a)131.00 (n/a)40.46 (n/a)

gelu_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.12 (+61.36%)0.10 (+65.53%)0.11 (+79.09%)0.04 (-5.63%)0.03 (+170.92%)208.20 (+5.95%)98.74 (-30.26%)71.90 (-44.13%)70.20 (-38.04%)61.20 (+79.92%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.06 (n/a)0.06 (n/a)0.04 (n/a)0.01 (n/a)196.50 (n/a)141.58 (n/a)128.70 (n/a)113.30 (n/a)34.01 (n/a)

gelu_8_cols_2_channels_2048_tile_128

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.10 (+64.20%)0.09 (+72.61%)0.10 (+77.79%)0.06 (+39.91%)0.02 (+119.55%)133.40 (-28.55%)93.44 (-40.99%)82.30 (-43.75%)81.90 (-39.11%)22.47 (-5.49%)
0f0da0b — 2026-01-14 16:31:000.06 (n/a)0.05 (n/a)0.06 (n/a)0.04 (n/a)0.01 (n/a)186.70 (n/a)158.34 (n/a)146.30 (n/a)134.50 (n/a)23.77 (n/a)

gemm_1792x896x1152_64x32x48_8cols_ccolmaj

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
1531781 — 2026-01-23 18:49:245.15 (+4.03%)4.90 (+9.86%)4.99 (+12.67%)4.27 (+4.36%)0.36 (+11.38%)2201.70 (-4.17%)1929.94 (-8.91%)1885.00 (-11.25%)1826.70 (-3.87%)154.05 (+4.53%)2025.19 (+4.03%)1925.76 (+9.86%)1962.51 (+12.67%)1680.20 (+4.36%)139.97 (+11.38%)
0f0da0b — 2026-01-14 16:31:004.95 (n/a)4.46 (n/a)4.43 (n/a)4.09 (n/a)0.32 (n/a)2297.60 (n/a)2118.80 (n/a)2123.90 (n/a)1900.20 (n/a)147.37 (n/a)1946.79 (n/a)1752.94 (n/a)1741.75 (n/a)1610.08 (n/a)125.67 (n/a)

gemm_192x384x64_48x96x16_4cols

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
1531781 — 2026-01-23 18:49:242.36 (+52.61%)1.85 (+39.10%)1.78 (+41.59%)1.32 (+14.31%)0.39 (+132.56%)167.80 (-12.51%)124.24 (-26.22%)124.40 (-29.36%)93.70 (-34.52%)28.11 (+35.82%)100.67 (+52.61%)78.94 (+39.10%)75.88 (+41.59%)56.25 (+14.31%)16.73 (+132.56%)
0f0da0b — 2026-01-14 16:31:001.55 (n/a)1.33 (n/a)1.26 (n/a)1.15 (n/a)0.17 (n/a)191.80 (n/a)168.40 (n/a)176.10 (n/a)143.10 (n/a)20.70 (n/a)65.96 (n/a)56.75 (n/a)53.59 (n/a)49.20 (n/a)7.20 (n/a)

gemm_192x384x64_48x96x16_4cols_bcolmaj_ccolmaj

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
1531781 — 2026-01-23 18:49:242.22 (+30.89%)2.06 (+54.21%)2.20 (+75.49%)1.60 (+35.74%)0.26 (+25.57%)138.20 (-26.33%)109.20 (-35.27%)100.50 (-43.03%)99.80 (-23.58%)16.54 (-28.10%)94.54 (+30.89%)87.77 (+54.21%)93.91 (+75.49%)68.30 (+35.74%)11.26 (+25.57%)
0f0da0b — 2026-01-14 16:31:001.69 (n/a)1.33 (n/a)1.25 (n/a)1.18 (n/a)0.21 (n/a)187.60 (n/a)168.70 (n/a)176.40 (n/a)130.60 (n/a)23.00 (n/a)72.23 (n/a)56.92 (n/a)53.51 (n/a)50.31 (n/a)8.97 (n/a)

gemm_2048x2048x2048_64x64x32_8_cols_0_bcolmaj_0_ccolmaj_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)GFLOP/s (max)GFLOP/s (mean)GFLOP/s (median)GFLOP/s (min)GFLOP/s (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3a053d3 — 2025-11-26 01:12:163.72 (-0.11%)3.55 (-2.75%)3.56 (-2.75%)3.36 (-5.33%)0.12 (+76.10%)2725.23 (-0.11%)2602.52 (-2.75%)2612.11 (-2.75%)2459.89 (-5.33%)86.27 (+76.10%)6984.00 (+5.63%)6608.60 (+2.91%)6577.00 (+2.83%)6304.00 (+0.11%)221.87 (+87.15%)
9e98049 — 2025-11-25 20:50:553.72 (n/a)3.65 (n/a)3.66 (n/a)3.54 (n/a)0.07 (n/a)2728.26 (n/a)2676.06 (n/a)2686.03 (n/a)2598.29 (n/a)48.99 (n/a)6612.00 (n/a)6422.00 (n/a)6396.00 (n/a)6297.00 (n/a)118.55 (n/a)

gemm_2048x2048x2048_64x64x32_8_cols_0_bcolmaj_1_ccolmaj_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)GFLOP/s (max)GFLOP/s (mean)GFLOP/s (median)GFLOP/s (min)GFLOP/s (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3a053d3 — 2025-11-26 01:12:163.40 (-3.14%)3.38 (-0.82%)3.38 (-0.82%)3.35 (+4.93%)0.02 (-84.51%)2493.81 (-3.14%)2475.27 (-0.82%)2479.06 (-0.82%)2453.92 (+4.93%)13.33 (-84.51%)7001.00 (-4.70%)6940.80 (+0.71%)6930.00 (+0.83%)6889.00 (+3.24%)37.43 (-84.76%)
9e98049 — 2025-11-25 20:50:553.51 (n/a)3.40 (n/a)3.41 (n/a)3.19 (n/a)0.12 (n/a)2574.53 (n/a)2495.72 (n/a)2499.62 (n/a)2338.67 (n/a)86.02 (n/a)7346.00 (n/a)6892.20 (n/a)6873.00 (n/a)6673.00 (n/a)245.57 (n/a)

gemm_2048x2048x2048_64x64x32_8_cols_1_bcolmaj_0_ccolmaj_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)GFLOP/s (max)GFLOP/s (mean)GFLOP/s (median)GFLOP/s (min)GFLOP/s (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3a053d3 — 2025-11-26 01:12:163.02 (+0.52%)2.87 (-2.05%)2.86 (-2.24%)2.70 (-6.16%)0.11 (+131.66%)2216.19 (+0.52%)2104.44 (-2.05%)2094.08 (-2.24%)1980.84 (-6.16%)80.25 (+131.67%)8673.00 (+6.56%)8175.60 (+2.21%)8204.00 (+2.29%)7752.00 (-0.51%)314.03 (+145.36%)
9e98049 — 2025-11-25 20:50:553.01 (n/a)2.93 (n/a)2.92 (n/a)2.88 (n/a)0.05 (n/a)2204.81 (n/a)2148.41 (n/a)2142.13 (n/a)2110.81 (n/a)34.64 (n/a)8139.00 (n/a)7998.60 (n/a)8020.00 (n/a)7792.00 (n/a)127.99 (n/a)

gemm_2048x2048x2048_64x64x64_1cols

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
1531781 — 2026-01-23 18:49:240.52 (-0.75%)0.52 (-0.07%)0.52 (+0.01%)0.52 (+0.22%)0.00 (-86.34%)48415.10 (-0.22%)48370.52 (+0.07%)48362.20 (-0.01%)48344.40 (+0.76%)29.40 (-86.25%)355.36 (-0.75%)355.17 (-0.07%)355.23 (+0.01%)354.85 (+0.22%)0.22 (-86.34%)
0f0da0b — 2026-01-14 16:31:000.52 (n/a)0.52 (n/a)0.52 (n/a)0.52 (n/a)0.00 (n/a)48520.80 (n/a)48335.84 (n/a)48365.40 (n/a)47981.00 (n/a)213.79 (n/a)358.06 (n/a)355.43 (n/a)355.21 (n/a)354.07 (n/a)1.58 (n/a)

gemm_2048x2048x2048_64x64x64_2_cols_0_bcolmaj_0_ccolmaj_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:491.04 (+0.94%)1.03 (+0.23%)1.03 (+0.09%)1.02 (+0.07%)0.00 (+197.51%)24576.00 (-0.07%)24485.82 (-0.23%)24530.40 (-0.09%)24266.40 (-0.94%)111.26 (+194.35%)
0a6c11c — 2025-12-04 00:39:101.03 (n/a)1.03 (n/a)1.03 (n/a)1.02 (n/a)0.00 (n/a)24594.30 (n/a)24542.14 (n/a)24551.90 (n/a)24495.50 (n/a)37.80 (n/a)

gemm_2048x2048x2048_64x64x64_2_cols_0_bcolmaj_0_ccolmaj_0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
a4b6ffe — 2026-01-05 19:24:051.04 (+0.28%)1.03 (-0.15%)1.03 (+0.05%)1.02 (-0.52%)0.01 (+42.72%)24719.80 (+0.52%)24512.42 (+0.16%)24529.20 (-0.05%)24092.60 (-0.28%)252.92 (+43.15%)713.08 (+0.28%)700.92 (-0.15%)700.38 (+0.05%)694.98 (-0.52%)7.30 (+42.71%)
50ad37c — 2025-12-29 20:35:531.04 (n/a)1.03 (n/a)1.03 (n/a)1.02 (n/a)0.01 (n/a)24592.30 (n/a)24473.70 (n/a)24540.90 (n/a)24160.80 (n/a)176.68 (n/a)711.06 (n/a)702.00 (n/a)700.05 (n/a)698.59 (n/a)5.12 (n/a)

gemm_2048x2048x2048_64x64x64_2_cols_0_bcolmaj_1_ccolmaj_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.99 (+1.69%)0.97 (+0.29%)0.97 (-0.04%)0.97 (-0.15%)0.01 (+434.92%)26025.30 (+0.15%)25839.34 (-0.29%)25916.70 (+0.04%)25446.00 (-1.67%)202.05 (+425.44%)
0a6c11c — 2025-12-04 00:39:100.97 (n/a)0.97 (n/a)0.97 (n/a)0.97 (n/a)0.00 (n/a)25986.90 (n/a)25913.52 (n/a)25905.70 (n/a)25877.00 (n/a)38.45 (n/a)

gemm_2048x2048x2048_64x64x64_2_cols_0_bcolmaj_1_ccolmaj_0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
a4b6ffe — 2026-01-05 19:24:050.97 (-0.01%)0.97 (+0.03%)0.97 (+0.04%)0.97 (+0.00%)0.00 (-6.43%)25927.90 (-0.00%)25903.20 (-0.03%)25898.60 (-0.04%)25892.80 (+0.01%)14.05 (-6.56%)663.50 (-0.01%)663.23 (+0.03%)663.35 (+0.04%)662.60 (+0.00%)0.36 (-6.43%)
50ad37c — 2025-12-29 20:35:530.97 (n/a)0.97 (n/a)0.97 (n/a)0.97 (n/a)0.00 (n/a)25928.10 (n/a)25910.14 (n/a)25908.10 (n/a)25891.20 (n/a)15.04 (n/a)663.54 (n/a)663.06 (n/a)663.11 (n/a)662.60 (n/a)0.38 (n/a)

gemm_2048x2048x2048_64x64x64_2_cols_1_bcolmaj_0_ccolmaj_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.92 (-1.02%)0.92 (-0.66%)0.92 (-0.10%)0.91 (-0.78%)0.01 (-4.54%)27751.40 (+0.79%)27480.74 (+0.66%)27354.80 (+0.10%)27320.50 (+1.03%)182.69 (-2.84%)
0a6c11c — 2025-12-04 00:39:100.93 (n/a)0.92 (n/a)0.92 (n/a)0.91 (n/a)0.01 (n/a)27534.80 (n/a)27299.32 (n/a)27326.70 (n/a)27042.30 (n/a)188.04 (n/a)

gemm_2048x2048x2048_64x64x64_2_cols_1_bcolmaj_0_ccolmaj_0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
a4b6ffe — 2026-01-05 19:24:050.92 (+0.11%)0.91 (-0.12%)0.91 (-0.21%)0.91 (-0.24%)0.00 (+30.34%)27723.50 (+0.24%)27548.94 (+0.12%)27522.90 (+0.21%)27403.60 (-0.11%)124.26 (+30.58%)626.92 (+0.11%)623.62 (-0.12%)624.20 (-0.21%)619.69 (-0.24%)2.81 (+30.34%)
50ad37c — 2025-12-29 20:35:530.92 (n/a)0.91 (n/a)0.92 (n/a)0.91 (n/a)0.00 (n/a)27657.10 (n/a)27515.58 (n/a)27464.70 (n/a)27434.60 (n/a)95.16 (n/a)626.21 (n/a)624.37 (n/a)625.52 (n/a)621.17 (n/a)2.16 (n/a)

gemm_2048x2048x2048_64x64x64_2cols_bcolmaj

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
1531781 — 2026-01-23 18:49:240.93 (+1.26%)0.91 (+0.29%)0.91 (+0.24%)0.90 (-0.34%)0.01 (+132.37%)27826.70 (+0.34%)27533.84 (-0.28%)27519.90 (-0.24%)27105.20 (-1.24%)272.35 (+129.91%)633.82 (+1.26%)624.00 (+0.29%)624.27 (+0.24%)617.39 (-0.34%)6.21 (+132.36%)
0f0da0b — 2026-01-14 16:31:000.92 (n/a)0.91 (n/a)0.91 (n/a)0.91 (n/a)0.00 (n/a)27731.40 (n/a)27612.36 (n/a)27585.70 (n/a)27446.50 (n/a)118.46 (n/a)625.94 (n/a)622.19 (n/a)622.78 (n/a)619.51 (n/a)2.67 (n/a)

gemm_2048x2048x2048_64x64x64_8_cols_0_bcolmaj_0_ccolmaj_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)GFLOP/s (max)GFLOP/s (mean)GFLOP/s (median)GFLOP/s (min)GFLOP/s (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
31ab24e — 2025-11-22 02:01:563.65 (-1.06%)3.59 (+0.83%)3.63 (-0.26%)3.45 (+2.66%)0.07 (-44.71%)2678.08 (-1.06%)2632.13 (+0.83%)2658.60 (-0.26%)2528.31 (+2.66%)53.86 (-44.71%)6795.00 (-2.59%)6529.80 (-0.92%)6462.00 (+0.26%)6415.00 (+1.07%)137.14 (-45.27%)
d9a64c5 — 2025-11-21 20:49:353.69 (n/a)3.56 (n/a)3.64 (n/a)3.36 (n/a)0.13 (n/a)2706.77 (n/a)2610.50 (n/a)2665.61 (n/a)2462.71 (n/a)97.41 (n/a)6976.00 (n/a)6590.40 (n/a)6445.00 (n/a)6347.00 (n/a)250.58 (n/a)

gemm_2048x2048x2048_64x64x64_8_cols_0_bcolmaj_0_ccolmaj_0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
d48746f — 2026-01-12 22:17:103.98 (+0.20%)3.84 (-2.95%)3.76 (-5.27%)3.72 (-4.97%)0.13 (+473.97%)6760.00 (+5.23%)6559.50 (+3.13%)6697.10 (+5.56%)6321.00 (-0.20%)218.68 (+499.40%)2717.91 (+0.20%)2621.44 (-2.95%)2565.27 (-5.27%)2541.40 (-4.97%)88.37 (+473.97%)
331dcca — 2026-01-08 18:34:103.97 (n/a)3.96 (n/a)3.97 (n/a)3.92 (n/a)0.02 (n/a)6423.80 (n/a)6360.32 (n/a)6344.30 (n/a)6333.90 (n/a)36.48 (n/a)2712.39 (n/a)2701.17 (n/a)2707.91 (n/a)2674.40 (n/a)15.40 (n/a)

gemm_2048x2048x2048_64x64x64_8_cols_0_bcolmaj_1_ccolmaj_0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
d48746f — 2026-01-12 22:17:103.75 (+0.28%)3.70 (+1.25%)3.71 (-0.21%)3.65 (+4.77%)0.04 (-66.90%)6901.80 (-4.56%)6795.46 (-1.30%)6790.20 (+0.21%)6715.90 (-0.28%)67.05 (-68.40%)2558.08 (+0.28%)2528.33 (+1.25%)2530.10 (-0.21%)2489.18 (+4.77%)24.81 (-66.90%)
331dcca — 2026-01-08 18:34:103.74 (n/a)3.66 (n/a)3.71 (n/a)3.48 (n/a)0.11 (n/a)7231.20 (n/a)6885.18 (n/a)6775.70 (n/a)6734.90 (n/a)212.21 (n/a)2550.87 (n/a)2497.04 (n/a)2535.50 (n/a)2375.78 (n/a)74.96 (n/a)

gemm_2048x2048x2048_64x64x64_8_cols_1_bcolmaj_0_ccolmaj_0_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
d48746f — 2026-01-12 22:17:103.62 (-0.77%)3.54 (+0.14%)3.60 (-0.04%)3.44 (+2.17%)0.09 (-25.77%)7307.40 (-2.13%)7103.46 (-0.18%)6982.50 (+0.04%)6948.20 (+0.77%)182.95 (-26.55%)2472.55 (-0.77%)2419.80 (+0.14%)2460.41 (-0.04%)2351.04 (+2.17%)61.76 (-25.77%)
331dcca — 2026-01-08 18:34:103.65 (n/a)3.54 (n/a)3.61 (n/a)3.37 (n/a)0.12 (n/a)7466.20 (n/a)7116.40 (n/a)6979.60 (n/a)6895.10 (n/a)249.09 (n/a)2491.61 (n/a)2416.45 (n/a)2461.45 (n/a)2301.00 (n/a)83.20 (n/a)

gemm_2048x2048x2048_64x64x64_8cols_bcolmaj_ccolmaj

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
1531781 — 2026-01-23 18:49:243.46 (+1.18%)3.42 (+2.60%)3.45 (+2.90%)3.29 (+2.92%)0.07 (-25.59%)7651.20 (-2.83%)7368.34 (-2.57%)7300.80 (-2.82%)7263.70 (-1.17%)162.53 (-28.22%)2365.17 (+1.18%)2332.46 (+2.60%)2353.14 (+2.90%)2245.39 (+2.92%)50.15 (-25.59%)
0f0da0b — 2026-01-14 16:31:003.42 (n/a)3.33 (n/a)3.35 (n/a)3.20 (n/a)0.10 (n/a)7874.40 (n/a)7562.58 (n/a)7512.80 (n/a)7349.50 (n/a)226.42 (n/a)2337.56 (n/a)2273.31 (n/a)2286.73 (n/a)2181.74 (n/a)67.40 (n/a)

gemm_384x1536x1792_32x48x64_4cols_bcolmaj

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
1531781 — 2026-01-23 18:49:244.35 (-0.80%)4.24 (+3.81%)4.33 (+9.58%)3.85 (-0.98%)0.22 (-4.90%)2094.90 (+0.99%)1905.64 (-3.69%)1862.00 (-8.75%)1852.00 (+0.81%)105.88 (-3.06%)1141.45 (-0.80%)1111.86 (+3.81%)1135.29 (+9.58%)1009.08 (-0.98%)57.51 (-4.90%)
0f0da0b — 2026-01-14 16:31:004.39 (n/a)4.08 (n/a)3.95 (n/a)3.89 (n/a)0.23 (n/a)2074.30 (n/a)1978.70 (n/a)2040.50 (n/a)1837.10 (n/a)109.22 (n/a)1150.67 (n/a)1071.02 (n/a)1036.00 (n/a)1019.11 (n/a)60.48 (n/a)

gemm_896x1792x640_32x64x80_8cols_ccolmaj

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
1531781 — 2026-01-23 18:49:246.80 (+27.89%)6.43 (+24.70%)6.70 (+30.77%)5.33 (+5.74%)0.62 (+474.69%)1247.10 (-5.44%)1042.70 (-19.16%)992.90 (-23.53%)978.30 (-21.81%)114.59 (+329.89%)2100.72 (+27.89%)1987.76 (+24.70%)2069.85 (+30.77%)1647.93 (+5.74%)190.83 (+474.69%)
0f0da0b — 2026-01-14 16:31:005.32 (n/a)5.16 (n/a)5.12 (n/a)5.04 (n/a)0.11 (n/a)1318.80 (n/a)1289.76 (n/a)1298.50 (n/a)1251.20 (n/a)26.66 (n/a)1642.57 (n/a)1594.05 (n/a)1582.78 (n/a)1558.44 (n/a)33.21 (n/a)

input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3cb43db — 2026-02-10 01:03:390.04 (n/a)0.03 (n/a)0.03 (n/a)0.01 (n/a)0.01 (n/a)559.60 (n/a)285.18 (n/a)252.60 (n/a)217.70 (n/a)85.28 (n/a)

input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048-group_size_32

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3cb43db — 2026-02-10 01:03:390.02 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)0.01 (n/a)744.10 (n/a)386.32 (n/a)287.50 (n/a)248.40 (n/a)209.55 (n/a)

input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048-weighted_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3cb43db — 2026-02-10 01:03:390.03 (n/a)0.03 (n/a)0.03 (n/a)0.02 (n/a)0.00 (n/a)361.70 (n/a)274.42 (n/a)253.20 (n/a)248.20 (n/a)48.94 (n/a)

input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3cb43db — 2026-02-10 01:03:390.04 (n/a)0.03 (n/a)0.03 (n/a)0.02 (n/a)0.01 (n/a)357.20 (n/a)264.56 (n/a)237.25 (n/a)219.10 (n/a)52.02 (n/a)

input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024-group_size_32

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3cb43db — 2026-02-10 01:03:390.02 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)0.00 (n/a)380.00 (n/a)278.08 (n/a)268.70 (n/a)231.00 (n/a)59.79 (n/a)

input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024-weighted_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3cb43db — 2026-02-10 01:03:390.04 (n/a)0.03 (n/a)0.03 (n/a)0.02 (n/a)0.01 (n/a)364.30 (n/a)270.96 (n/a)250.60 (n/a)231.30 (n/a)53.57 (n/a)

input_length_2048-num_aie_columns_1-num_channels_2-tile_size_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3cb43db — 2026-02-10 01:03:390.06 (n/a)0.04 (n/a)0.05 (n/a)0.01 (n/a)0.01 (n/a)829.20 (n/a)322.22 (n/a)240.05 (n/a)223.40 (n/a)184.50 (n/a)

input_length_2048-num_aie_columns_1-num_channels_2-tile_size_2048-scalar_factor_3.0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3cb43db — 2026-02-10 01:03:390.05 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)347.10 (n/a)271.02 (n/a)248.90 (n/a)242.90 (n/a)43.79 (n/a)

input_length_2048-num_aie_columns_1-num_channels_2-tile_size_2048-weighted_True

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3cb43db — 2026-02-10 01:03:390.05 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.00 (n/a)313.50 (n/a)269.12 (n/a)254.60 (n/a)251.80 (n/a)26.33 (n/a)

input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3cb43db — 2026-02-10 01:03:390.04 (n/a)0.03 (n/a)0.03 (n/a)0.01 (n/a)0.01 (n/a)634.20 (n/a)276.63 (n/a)244.25 (n/a)227.00 (n/a)80.23 (n/a)

input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024-group_size_32

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3cb43db — 2026-02-10 01:03:390.02 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)0.00 (n/a)455.40 (n/a)291.12 (n/a)253.00 (n/a)243.40 (n/a)92.09 (n/a)

input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024-weighted_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3cb43db — 2026-02-10 01:03:390.03 (n/a)0.03 (n/a)0.03 (n/a)0.02 (n/a)0.00 (n/a)356.90 (n/a)283.90 (n/a)264.40 (n/a)245.30 (n/a)43.72 (n/a)

input_length_2048-num_aie_columns_2-num_channels_2-tile_size_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3cb43db — 2026-02-10 01:03:390.06 (n/a)0.05 (n/a)0.05 (n/a)0.02 (n/a)0.01 (n/a)667.30 (n/a)314.66 (n/a)240.10 (n/a)217.90 (n/a)165.14 (n/a)

input_length_2048-num_aie_columns_2-num_channels_2-tile_size_1024-scalar_factor_3.0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3cb43db — 2026-02-10 01:03:390.05 (n/a)0.05 (n/a)0.05 (n/a)0.02 (n/a)0.01 (n/a)523.10 (n/a)299.58 (n/a)244.10 (n/a)232.90 (n/a)125.24 (n/a)

input_length_2048-num_aie_columns_2-num_channels_2-tile_size_1024-weighted_True

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3cb43db — 2026-02-10 01:03:390.04 (n/a)0.04 (n/a)0.04 (n/a)0.03 (n/a)0.01 (n/a)346.70 (n/a)263.14 (n/a)243.40 (n/a)229.60 (n/a)48.57 (n/a)

input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3cb43db — 2026-02-10 01:03:390.04 (n/a)0.03 (n/a)0.03 (n/a)0.01 (n/a)0.01 (n/a)561.70 (n/a)318.76 (n/a)265.60 (n/a)230.30 (n/a)110.68 (n/a)

input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512-group_size_32

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3cb43db — 2026-02-10 01:03:390.02 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)0.00 (n/a)405.80 (n/a)277.76 (n/a)251.10 (n/a)230.10 (n/a)72.21 (n/a)

input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512-weighted_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3cb43db — 2026-02-10 01:03:390.03 (n/a)0.03 (n/a)0.03 (n/a)0.02 (n/a)0.01 (n/a)421.00 (n/a)289.06 (n/a)255.60 (n/a)250.30 (n/a)74.07 (n/a)

input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3cb43db — 2026-02-10 01:03:390.04 (n/a)0.03 (n/a)0.03 (n/a)0.02 (n/a)0.01 (n/a)456.60 (n/a)275.53 (n/a)246.45 (n/a)226.70 (n/a)61.93 (n/a)

input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512-group_size_32

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3cb43db — 2026-02-10 01:03:390.02 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)0.00 (n/a)380.80 (n/a)274.98 (n/a)253.40 (n/a)242.90 (n/a)59.37 (n/a)

input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512-weighted_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3cb43db — 2026-02-10 01:03:390.03 (n/a)0.03 (n/a)0.03 (n/a)0.02 (n/a)0.01 (n/a)418.70 (n/a)318.26 (n/a)294.20 (n/a)246.80 (n/a)70.17 (n/a)

input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3cb43db — 2026-02-10 01:03:390.04 (n/a)0.03 (n/a)0.03 (n/a)0.01 (n/a)0.01 (n/a)651.60 (n/a)307.56 (n/a)249.65 (n/a)233.70 (n/a)127.67 (n/a)

input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256-group_size_32

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3cb43db — 2026-02-10 01:03:390.02 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)0.01 (n/a)851.40 (n/a)375.90 (n/a)258.70 (n/a)250.60 (n/a)265.87 (n/a)

input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256-weighted_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3cb43db — 2026-02-10 01:03:390.03 (n/a)0.03 (n/a)0.03 (n/a)0.01 (n/a)0.01 (n/a)715.30 (n/a)371.22 (n/a)263.70 (n/a)250.10 (n/a)197.95 (n/a)

input_length_2048-num_aie_columns_4-num_channels_2-tile_size_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3cb43db — 2026-02-10 01:03:390.05 (n/a)0.05 (n/a)0.05 (n/a)0.02 (n/a)0.01 (n/a)715.50 (n/a)296.16 (n/a)243.20 (n/a)224.60 (n/a)151.42 (n/a)

input_length_2048-num_aie_columns_4-num_channels_2-tile_size_512-scalar_factor_3.0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3cb43db — 2026-02-10 01:03:390.05 (n/a)0.05 (n/a)0.05 (n/a)0.03 (n/a)0.01 (n/a)351.40 (n/a)273.22 (n/a)254.90 (n/a)249.40 (n/a)43.78 (n/a)

input_length_2048-num_aie_columns_4-num_channels_2-tile_size_512-weighted_True

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3cb43db — 2026-02-10 01:03:390.04 (n/a)0.03 (n/a)0.04 (n/a)0.03 (n/a)0.01 (n/a)355.70 (n/a)270.72 (n/a)249.10 (n/a)239.10 (n/a)48.62 (n/a)

input_length_2048-num_aie_columns_8-num_channels_1-tile_size_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3cb43db — 2026-02-10 01:03:390.04 (n/a)0.03 (n/a)0.03 (n/a)0.01 (n/a)0.01 (n/a)717.60 (n/a)286.43 (n/a)248.20 (n/a)225.10 (n/a)109.36 (n/a)

input_length_2048-num_aie_columns_8-num_channels_1-tile_size_256-group_size_32

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3cb43db — 2026-02-10 01:03:390.02 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)0.00 (n/a)373.10 (n/a)278.38 (n/a)272.00 (n/a)227.00 (n/a)59.27 (n/a)

input_length_2048-num_aie_columns_8-num_channels_1-tile_size_256-weighted_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3cb43db — 2026-02-10 01:03:390.03 (n/a)0.03 (n/a)0.03 (n/a)0.02 (n/a)0.00 (n/a)361.20 (n/a)290.88 (n/a)290.40 (n/a)250.60 (n/a)45.51 (n/a)

input_length_2048-num_aie_columns_8-num_channels_2-tile_size_128

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3cb43db — 2026-02-10 01:03:390.03 (n/a)0.03 (n/a)0.03 (n/a)0.01 (n/a)0.01 (n/a)722.40 (n/a)363.25 (n/a)264.20 (n/a)242.70 (n/a)172.70 (n/a)

input_length_2048-num_aie_columns_8-num_channels_2-tile_size_128-group_size_32

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3cb43db — 2026-02-10 01:03:390.02 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)0.01 (n/a)660.20 (n/a)350.14 (n/a)273.70 (n/a)249.90 (n/a)174.20 (n/a)

input_length_2048-num_aie_columns_8-num_channels_2-tile_size_128-weighted_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3cb43db — 2026-02-10 01:03:390.03 (n/a)0.03 (n/a)0.03 (n/a)0.01 (n/a)0.01 (n/a)987.80 (n/a)400.64 (n/a)252.50 (n/a)247.70 (n/a)328.32 (n/a)

input_length_2048-num_aie_columns_8-num_channels_2-tile_size_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3cb43db — 2026-02-10 01:03:390.05 (n/a)0.04 (n/a)0.05 (n/a)0.02 (n/a)0.01 (n/a)681.20 (n/a)330.94 (n/a)255.80 (n/a)244.90 (n/a)149.84 (n/a)

input_length_2048-num_aie_columns_8-num_channels_2-tile_size_256-scalar_factor_3.0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3cb43db — 2026-02-10 01:03:390.05 (n/a)0.04 (n/a)0.05 (n/a)0.02 (n/a)0.01 (n/a)647.10 (n/a)325.12 (n/a)239.10 (n/a)227.90 (n/a)180.94 (n/a)

input_length_2048-num_aie_columns_8-num_channels_2-tile_size_256-weighted_True

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3cb43db — 2026-02-10 01:03:390.03 (n/a)0.03 (n/a)0.03 (n/a)0.02 (n/a)0.01 (n/a)493.50 (n/a)305.26 (n/a)258.40 (n/a)254.60 (n/a)105.28 (n/a)

input_length_2048-num_cores_1-num_channels_1-bypass_False-tile_size_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3cb43db — 2026-02-10 01:03:390.03 (n/a)0.03 (n/a)0.03 (n/a)0.02 (n/a)0.00 (n/a)340.90 (n/a)275.36 (n/a)273.20 (n/a)239.80 (n/a)41.42 (n/a)

input_length_2048-num_cores_16-num_channels_2-bypass_False-tile_size_128

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3cb43db — 2026-02-10 01:03:390.03 (n/a)0.03 (n/a)0.03 (n/a)0.01 (n/a)0.01 (n/a)730.80 (n/a)384.00 (n/a)270.60 (n/a)257.70 (n/a)203.01 (n/a)

input_length_2048-num_cores_2-num_channels_1-bypass_False-tile_size_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3cb43db — 2026-02-10 01:03:390.04 (n/a)0.03 (n/a)0.03 (n/a)0.02 (n/a)0.00 (n/a)334.50 (n/a)257.58 (n/a)240.10 (n/a)224.80 (n/a)43.87 (n/a)

input_length_2048-num_cores_2-num_channels_2-bypass_False-tile_size_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3cb43db — 2026-02-10 01:03:390.04 (n/a)0.03 (n/a)0.04 (n/a)0.02 (n/a)0.01 (n/a)337.60 (n/a)248.68 (n/a)230.10 (n/a)213.60 (n/a)50.67 (n/a)

input_length_2048-num_cores_4-num_channels_1-bypass_False-tile_size_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3cb43db — 2026-02-10 01:03:390.04 (n/a)0.03 (n/a)0.04 (n/a)0.02 (n/a)0.01 (n/a)378.60 (n/a)265.18 (n/a)230.30 (n/a)225.10 (n/a)65.57 (n/a)

input_length_2048-num_cores_4-num_channels_2-bypass_False-tile_size_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3cb43db — 2026-02-10 01:03:390.04 (n/a)0.03 (n/a)0.03 (n/a)0.02 (n/a)0.01 (n/a)448.90 (n/a)283.70 (n/a)244.10 (n/a)224.40 (n/a)93.11 (n/a)

input_length_2048-num_cores_8-num_channels_1-bypass_False-tile_size_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3cb43db — 2026-02-10 01:03:390.04 (n/a)0.03 (n/a)0.03 (n/a)0.01 (n/a)0.01 (n/a)655.90 (n/a)322.62 (n/a)236.50 (n/a)233.20 (n/a)186.47 (n/a)

input_length_2048-num_cores_8-num_channels_2-bypass_False-tile_size_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3cb43db — 2026-02-10 01:03:390.03 (n/a)0.03 (n/a)0.03 (n/a)0.02 (n/a)0.01 (n/a)432.50 (n/a)293.02 (n/a)253.00 (n/a)243.70 (n/a)80.44 (n/a)

input_length_32768-num_aie_columns_2-num_channels_2-tile_size_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3cb43db — 2026-02-10 01:03:390.57 (n/a)0.43 (n/a)0.48 (n/a)0.28 (n/a)0.13 (n/a)471.10 (n/a)329.88 (n/a)273.50 (n/a)228.90 (n/a)106.23 (n/a)

input_length_32768-num_aie_columns_2-num_channels_2-tile_size_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3cb43db — 2026-02-10 01:03:390.53 (n/a)0.48 (n/a)0.50 (n/a)0.37 (n/a)0.07 (n/a)355.70 (n/a)279.52 (n/a)261.50 (n/a)246.80 (n/a)44.31 (n/a)

input_length_32768-num_aie_columns_4-num_channels_4-tile_size_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3cb43db — 2026-02-10 01:03:390.52 (n/a)0.44 (n/a)0.50 (n/a)0.19 (n/a)0.14 (n/a)687.40 (n/a)344.52 (n/a)263.70 (n/a)250.80 (n/a)191.83 (n/a)

layer_norm_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.11 (+88.61%)0.09 (+61.18%)0.08 (+46.00%)0.07 (+61.81%)0.02 (+189.47%)119.70 (-38.20%)97.44 (-36.14%)100.50 (-31.54%)71.50 (-46.96%)22.67 (-5.36%)
0f0da0b — 2026-01-14 16:31:000.06 (n/a)0.05 (n/a)0.06 (n/a)0.04 (n/a)0.01 (n/a)193.70 (n/a)152.58 (n/a)146.80 (n/a)134.80 (n/a)23.96 (n/a)

layer_norm_1_cols_2_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.11 (+73.83%)0.09 (+50.07%)0.08 (+46.18%)0.06 (+7.28%)0.02 (+683.83%)138.10 (-6.82%)98.78 (-30.35%)97.10 (-31.57%)76.40 (-42.47%)24.88 (+313.86%)
0f0da0b — 2026-01-14 16:31:000.06 (n/a)0.06 (n/a)0.06 (n/a)0.06 (n/a)0.00 (n/a)148.20 (n/a)141.82 (n/a)141.90 (n/a)132.80 (n/a)6.01 (n/a)

layer_norm_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.12 (+58.00%)0.09 (+35.21%)0.09 (+33.88%)0.06 (+12.88%)0.02 (+142.40%)140.10 (-11.44%)98.38 (-23.32%)94.40 (-25.26%)68.10 (-36.71%)26.11 (+36.70%)
0f0da0b — 2026-01-14 16:31:000.08 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)158.20 (n/a)128.30 (n/a)126.30 (n/a)107.60 (n/a)19.10 (n/a)

layer_norm_2_cols_2_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.13 (+115.82%)0.09 (+79.97%)0.09 (+57.15%)0.06 (+84.94%)0.03 (+142.79%)130.40 (-45.94%)91.90 (-43.54%)93.50 (-36.39%)62.00 (-53.66%)25.59 (-42.10%)
0f0da0b — 2026-01-14 16:31:000.06 (n/a)0.05 (n/a)0.06 (n/a)0.03 (n/a)0.01 (n/a)241.20 (n/a)162.78 (n/a)147.00 (n/a)133.80 (n/a)44.20 (n/a)

layer_norm_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.11 (+63.95%)0.09 (+43.09%)0.11 (+55.71%)0.06 (-6.72%)0.02 (+806.48%)141.00 (+7.22%)91.54 (-25.97%)77.90 (-35.78%)73.00 (-39.07%)28.43 (+505.85%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.07 (n/a)0.07 (n/a)0.06 (n/a)0.00 (n/a)131.50 (n/a)123.66 (n/a)121.30 (n/a)119.80 (n/a)4.69 (n/a)

layer_norm_4_cols_2_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.11 (+76.69%)0.09 (+61.27%)0.09 (+70.26%)0.06 (+36.41%)0.02 (+195.74%)148.40 (-26.72%)103.14 (-34.74%)87.80 (-41.27%)73.70 (-43.44%)32.30 (+19.79%)
0f0da0b — 2026-01-14 16:31:000.06 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)202.50 (n/a)158.04 (n/a)149.50 (n/a)130.30 (n/a)26.96 (n/a)

layer_norm_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.12 (+75.09%)0.10 (+78.44%)0.11 (+94.76%)0.05 (+12.92%)0.03 (+213.89%)153.90 (-11.45%)90.36 (-39.81%)77.40 (-48.67%)67.30 (-42.87%)35.86 (+72.31%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.06 (n/a)0.05 (n/a)0.05 (n/a)0.01 (n/a)173.80 (n/a)150.12 (n/a)150.80 (n/a)117.80 (n/a)20.81 (n/a)

layer_norm_8_cols_2_channels_2048_tile_128

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.09 (+88.05%)0.08 (+72.40%)0.08 (+83.22%)0.06 (+32.78%)0.01 (+723.36%)140.30 (-24.69%)106.50 (-40.67%)97.60 (-45.44%)92.40 (-46.80%)19.84 (+234.33%)
0f0da0b — 2026-01-14 16:31:000.05 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.00 (n/a)186.30 (n/a)179.50 (n/a)178.90 (n/a)173.70 (n/a)5.93 (n/a)

matrix_vector_mul_128x128_32_1col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:490.37 (+17.13%)0.30 (+24.87%)0.30 (+38.03%)0.23 (+12.47%)0.06 (+35.23%)
0a6c11c — 2025-12-04 00:39:100.32 (n/a)0.24 (n/a)0.22 (n/a)0.20 (n/a)0.04 (n/a)

matrix_vector_mul_128x128_32_1col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:100.26 (-1.81%)0.24 (+4.06%)0.24 (+0.52%)0.21 (+16.97%)0.02 (-48.16%)0.26 (-1.81%)0.24 (+4.06%)0.24 (+0.52%)0.20 (+16.97%)0.02 (-48.16%)
a4b6ffe — 2026-01-05 19:24:050.26 (n/a)0.23 (n/a)0.24 (n/a)0.18 (n/a)0.04 (n/a)0.26 (n/a)0.23 (n/a)0.24 (n/a)0.17 (n/a)0.04 (n/a)

matrix_vector_mul_128x128_32tsi_128tso_1col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
1531781 — 2026-01-23 18:49:240.45 (+63.20%)0.35 (+43.38%)0.35 (+38.29%)0.22 (+12.34%)0.08 (+156.17%)0.44 (+63.20%)0.34 (+43.38%)0.34 (+38.29%)0.22 (+12.34%)0.08 (+156.17%)
0f0da0b — 2026-01-14 16:31:000.27 (n/a)0.24 (n/a)0.25 (n/a)0.20 (n/a)0.03 (n/a)0.27 (n/a)0.24 (n/a)0.25 (n/a)0.19 (n/a)0.03 (n/a)

matrix_vector_mul_2048x8192_1_1col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:4913.57 (+3.47%)13.15 (+1.09%)13.10 (+0.16%)12.52 (-1.58%)0.37 (+153.19%)
0a6c11c — 2025-12-04 00:39:1013.11 (n/a)13.01 (n/a)13.07 (n/a)12.72 (n/a)0.15 (n/a)

matrix_vector_mul_2048x8192_1_1col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:1013.43 (-0.07%)13.30 (+4.62%)13.27 (+1.29%)13.16 (+16.70%)0.12 (-86.69%)13.42 (-0.07%)13.29 (+4.62%)13.26 (+1.29%)13.16 (+16.70%)0.12 (-86.69%)
a4b6ffe — 2026-01-05 19:24:0513.44 (n/a)12.71 (n/a)13.10 (n/a)11.28 (n/a)0.87 (n/a)13.43 (n/a)12.70 (n/a)13.09 (n/a)11.27 (n/a)0.87 (n/a)

matrix_vector_mul_2048x8192_1_2col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:4925.06 (+2.47%)23.71 (+1.81%)24.13 (+0.50%)20.90 (+5.41%)1.48 (-15.60%)
0a6c11c — 2025-12-04 00:39:1024.46 (n/a)23.29 (n/a)24.01 (n/a)19.83 (n/a)1.75 (n/a)

matrix_vector_mul_2048x8192_1_2col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:1024.46 (-0.60%)21.07 (-8.03%)19.75 (-17.83%)18.08 (+1.18%)3.03 (+7.07%)24.44 (-0.60%)21.06 (-8.03%)19.74 (-17.83%)18.07 (+1.18%)3.03 (+7.07%)
a4b6ffe — 2026-01-05 19:24:0524.60 (n/a)22.91 (n/a)24.04 (n/a)17.87 (n/a)2.83 (n/a)24.59 (n/a)22.90 (n/a)24.02 (n/a)17.86 (n/a)2.83 (n/a)

matrix_vector_mul_2048x8192_1_4col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:4942.05 (+1.54%)40.50 (+12.60%)40.49 (+2.36%)38.98 (+97.53%)0.97 (-88.09%)
0a6c11c — 2025-12-04 00:39:1041.41 (n/a)35.97 (n/a)39.55 (n/a)19.73 (n/a)8.15 (n/a)

matrix_vector_mul_2048x8192_1_4col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:1042.58 (+1.51%)41.37 (+14.29%)41.25 (+4.89%)40.33 (+91.42%)0.92 (-89.18%)42.56 (+1.51%)41.34 (+14.29%)41.23 (+4.89%)40.30 (+91.42%)0.92 (-89.18%)
a4b6ffe — 2026-01-05 19:24:0541.95 (n/a)36.20 (n/a)39.33 (n/a)21.07 (n/a)8.54 (n/a)41.92 (n/a)36.17 (n/a)39.31 (n/a)21.05 (n/a)8.53 (n/a)

matrix_vector_mul_2048x8192_1_8col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:4946.62 (+3.11%)44.82 (+0.17%)44.78 (+0.17%)42.60 (-3.80%)1.60 (+367.72%)
0a6c11c — 2025-12-04 00:39:1045.21 (n/a)44.75 (n/a)44.70 (n/a)44.29 (n/a)0.34 (n/a)

matrix_vector_mul_2048x8192_1_8col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:1046.04 (+4.14%)45.06 (+3.47%)44.55 (+2.76%)44.30 (+2.84%)0.87 (+66.58%)46.01 (+4.14%)45.03 (+3.47%)44.52 (+2.76%)44.27 (+2.84%)0.87 (+66.58%)
a4b6ffe — 2026-01-05 19:24:0544.21 (n/a)43.55 (n/a)43.35 (n/a)43.07 (n/a)0.52 (n/a)44.18 (n/a)43.52 (n/a)43.32 (n/a)43.05 (n/a)0.52 (n/a)

matrix_vector_mul_2048x8192_1tsi_1024tso_2col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
1531781 — 2026-01-23 18:49:2424.80 (+0.16%)24.59 (-0.08%)24.55 (-0.22%)24.40 (+0.04%)0.18 (+11.75%)24.79 (+0.16%)24.58 (-0.08%)24.53 (-0.22%)24.38 (+0.04%)0.18 (+11.75%)
0f0da0b — 2026-01-14 16:31:0024.76 (n/a)24.61 (n/a)24.60 (n/a)24.39 (n/a)0.16 (n/a)24.75 (n/a)24.60 (n/a)24.58 (n/a)24.37 (n/a)0.16 (n/a)

matrix_vector_mul_2048x8192_1tsi_2048tso_1col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
1531781 — 2026-01-23 18:49:2413.54 (+0.77%)13.34 (+3.03%)13.45 (+0.75%)13.10 (+15.45%)0.21 (-77.02%)13.53 (+0.77%)13.33 (+3.03%)13.44 (+0.75%)13.09 (+15.45%)0.21 (-77.02%)
0f0da0b — 2026-01-14 16:31:0013.44 (n/a)12.95 (n/a)13.35 (n/a)11.35 (n/a)0.90 (n/a)13.43 (n/a)12.94 (n/a)13.34 (n/a)11.34 (n/a)0.90 (n/a)

matrix_vector_mul_2048x8192_1tsi_256tso_8col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
1531781 — 2026-01-23 18:49:2446.81 (+1.97%)45.91 (+2.21%)46.38 (+2.98%)44.81 (+2.36%)0.89 (-7.11%)46.79 (+1.97%)45.88 (+2.21%)46.35 (+2.98%)44.79 (+2.36%)0.89 (-7.11%)
0f0da0b — 2026-01-14 16:31:0045.91 (n/a)44.92 (n/a)45.04 (n/a)43.78 (n/a)0.96 (n/a)45.88 (n/a)44.89 (n/a)45.01 (n/a)43.75 (n/a)0.96 (n/a)

matrix_vector_mul_2048x8192_1tsi_512tso_4col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
1531781 — 2026-01-23 18:49:2441.49 (-0.92%)40.81 (+0.09%)41.13 (+0.38%)39.95 (+0.86%)0.65 (-24.40%)41.46 (-0.92%)40.79 (+0.09%)41.10 (+0.38%)39.93 (+0.86%)0.65 (-24.40%)
0f0da0b — 2026-01-14 16:31:0041.88 (n/a)40.78 (n/a)40.97 (n/a)39.61 (n/a)0.86 (n/a)41.85 (n/a)40.75 (n/a)40.95 (n/a)39.59 (n/a)0.86 (n/a)

matrix_vector_mul_8192x2048_4_1col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:4913.29 (+1.63%)13.19 (+6.08%)13.19 (+1.17%)13.09 (+26.86%)0.08 (-92.94%)
0a6c11c — 2025-12-04 00:39:1013.08 (n/a)12.44 (n/a)13.03 (n/a)10.32 (n/a)1.07 (n/a)

matrix_vector_mul_8192x2048_4_1col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:1013.68 (+3.42%)13.30 (+8.54%)13.38 (+7.25%)12.72 (+14.96%)0.35 (-66.09%)13.67 (+3.42%)13.29 (+8.54%)13.38 (+7.25%)12.71 (+14.96%)0.35 (-66.09%)
a4b6ffe — 2026-01-05 19:24:0513.23 (n/a)12.25 (n/a)12.48 (n/a)11.07 (n/a)1.04 (n/a)13.22 (n/a)12.24 (n/a)12.47 (n/a)11.06 (n/a)1.04 (n/a)

matrix_vector_mul_8192x2048_4_2col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:4925.58 (+4.24%)24.23 (+6.59%)24.03 (+0.29%)23.58 (+37.29%)0.70 (-74.97%)
0a6c11c — 2025-12-04 00:39:1024.54 (n/a)22.73 (n/a)23.96 (n/a)17.18 (n/a)2.79 (n/a)

matrix_vector_mul_8192x2048_4_2col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:1025.08 (+1.32%)24.68 (+13.42%)24.61 (+3.26%)24.47 (+48.32%)0.24 (-93.56%)25.07 (+1.32%)24.67 (+13.42%)24.59 (+3.26%)24.46 (+48.32%)0.24 (-93.56%)
a4b6ffe — 2026-01-05 19:24:0524.76 (n/a)21.76 (n/a)23.83 (n/a)16.50 (n/a)3.74 (n/a)24.74 (n/a)21.75 (n/a)23.82 (n/a)16.49 (n/a)3.73 (n/a)

matrix_vector_mul_8192x2048_4_4col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:4940.94 (-2.19%)39.74 (-1.10%)39.76 (-0.60%)38.17 (-2.14%)1.04 (+4.39%)
0a6c11c — 2025-12-04 00:39:1041.86 (n/a)40.18 (n/a)40.00 (n/a)39.00 (n/a)1.00 (n/a)

matrix_vector_mul_8192x2048_4_4col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:1040.74 (-1.09%)37.16 (-7.15%)40.25 (+1.47%)24.51 (-37.44%)7.07 (+736.13%)40.72 (-1.09%)37.13 (-7.15%)40.23 (+1.47%)24.50 (-37.44%)7.07 (+736.13%)
a4b6ffe — 2026-01-05 19:24:0541.19 (n/a)40.02 (n/a)39.67 (n/a)39.18 (n/a)0.85 (n/a)41.17 (n/a)39.99 (n/a)39.64 (n/a)39.16 (n/a)0.85 (n/a)

matrix_vector_mul_8192x2048_4_8col

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)
130b6ea — 2025-12-05 21:15:4946.71 (-1.04%)44.91 (-2.34%)44.71 (-2.98%)43.80 (-1.44%)0.96 (+6.93%)
0a6c11c — 2025-12-04 00:39:1047.20 (n/a)45.98 (n/a)46.08 (n/a)44.44 (n/a)0.90 (n/a)

matrix_vector_mul_8192x2048_4_8col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
331dcca — 2026-01-08 18:34:1046.28 (+5.68%)44.89 (+14.59%)44.74 (+4.26%)44.20 (+62.80%)0.84 (-87.96%)46.25 (+5.68%)44.86 (+14.59%)44.71 (+4.26%)44.17 (+62.80%)0.84 (-87.96%)
a4b6ffe — 2026-01-05 19:24:0543.79 (n/a)39.17 (n/a)42.91 (n/a)27.15 (n/a)6.98 (n/a)43.77 (n/a)39.15 (n/a)42.88 (n/a)27.13 (n/a)6.98 (n/a)

matrix_vector_mul_8192x2048_4tsi_1024tso_1col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
1531781 — 2026-01-23 18:49:2413.46 (+0.53%)13.33 (+2.85%)13.32 (+1.17%)13.20 (+10.91%)0.11 (-81.73%)13.45 (+0.53%)13.32 (+2.85%)13.31 (+1.17%)13.19 (+10.91%)0.11 (-81.73%)
0f0da0b — 2026-01-14 16:31:0013.39 (n/a)12.96 (n/a)13.17 (n/a)11.90 (n/a)0.62 (n/a)13.38 (n/a)12.95 (n/a)13.16 (n/a)11.89 (n/a)0.62 (n/a)

matrix_vector_mul_8192x2048_4tsi_1024tso_2col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
1531781 — 2026-01-23 18:49:2425.27 (+0.55%)24.75 (-0.75%)24.61 (-1.28%)24.55 (-0.37%)0.30 (+52.22%)25.25 (+0.55%)24.74 (-0.75%)24.59 (-1.28%)24.53 (-0.37%)0.30 (+52.22%)
0f0da0b — 2026-01-14 16:31:0025.13 (n/a)24.94 (n/a)24.93 (n/a)24.64 (n/a)0.20 (n/a)25.11 (n/a)24.92 (n/a)24.91 (n/a)24.62 (n/a)0.20 (n/a)

matrix_vector_mul_8192x2048_4tsi_1024tso_4col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
1531781 — 2026-01-23 18:49:2440.86 (-4.37%)40.01 (-4.03%)40.43 (-3.32%)38.86 (-2.46%)0.89 (-24.84%)40.84 (-4.37%)39.98 (-4.03%)40.41 (-3.32%)38.84 (-2.46%)0.89 (-24.84%)
0f0da0b — 2026-01-14 16:31:0042.73 (n/a)41.69 (n/a)41.82 (n/a)39.85 (n/a)1.18 (n/a)42.70 (n/a)41.66 (n/a)41.80 (n/a)39.82 (n/a)1.18 (n/a)

matrix_vector_mul_8192x2048_4tsi_1024tso_8col0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
1531781 — 2026-01-23 18:49:2446.08 (-1.82%)45.11 (-1.49%)45.49 (-0.25%)43.89 (-1.05%)0.97 (-6.46%)46.05 (-1.82%)45.08 (-1.49%)45.46 (-0.25%)43.87 (-1.05%)0.97 (-6.46%)
0f0da0b — 2026-01-14 16:31:0046.94 (n/a)45.79 (n/a)45.60 (n/a)44.36 (n/a)1.03 (n/a)46.91 (n/a)45.76 (n/a)45.58 (n/a)44.33 (n/a)1.03 (n/a)

mem_copy_16_cores_2_chans_2048_tile_128_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.07 (+22.03%)0.06 (+27.86%)0.06 (+35.49%)0.05 (+15.00%)0.01 (+49.69%)173.30 (-13.00%)137.94 (-21.32%)133.60 (-26.19%)123.40 (-18.06%)18.40 (+8.53%)
0a6c11c — 2025-12-04 00:39:100.05 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.00 (n/a)199.20 (n/a)175.32 (n/a)181.00 (n/a)150.60 (n/a)16.96 (n/a)

mem_copy_16_cores_2_chans_2048_tile_128_False0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.10 (+63.55%)0.08 (+85.69%)0.09 (+119.34%)0.04 (+23.15%)0.02 (+126.16%)195.30 (-18.79%)111.62 (-42.82%)86.60 (-54.42%)84.60 (-38.87%)47.65 (+9.68%)
0f0da0b — 2026-01-14 16:31:000.06 (n/a)0.04 (n/a)0.04 (n/a)0.03 (n/a)0.01 (n/a)240.50 (n/a)195.20 (n/a)190.00 (n/a)138.40 (n/a)43.45 (n/a)

mem_copy_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
003b49d — 2025-11-14 19:30:540.05 (-16.99%)0.05 (-12.47%)0.05 (-7.32%)0.04 (-11.17%)0.00 (-29.61%)197.00 (+12.57%)171.00 (+13.70%)164.00 (+7.89%)153.00 (+20.47%)17.49 (-4.32%)
0999a9a — 2025-11-13 19:24:080.06 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)175.00 (n/a)150.40 (n/a)152.00 (n/a)127.00 (n/a)18.28 (n/a)

mem_copy_1_cols_2_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
003b49d — 2025-11-14 19:30:540.06 (+5.56%)0.05 (-1.59%)0.05 (-1.39%)0.03 (-14.07%)0.01 (+25.03%)263.00 (+16.37%)166.00 (+4.53%)144.00 (+1.41%)126.00 (-5.26%)49.38 (+43.17%)
0999a9a — 2025-11-13 19:24:080.06 (n/a)0.05 (n/a)0.05 (n/a)0.03 (n/a)0.01 (n/a)226.00 (n/a)158.80 (n/a)142.00 (n/a)133.00 (n/a)34.49 (n/a)

mem_copy_1_cores_1_chans_2048_tile_2048_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.08 (+10.93%)0.07 (+7.79%)0.07 (+3.55%)0.05 (-1.73%)0.01 (+36.73%)150.20 (+1.76%)121.68 (-6.36%)125.60 (-3.38%)97.90 (-9.85%)18.56 (+21.98%)
0a6c11c — 2025-12-04 00:39:100.08 (n/a)0.06 (n/a)0.06 (n/a)0.06 (n/a)0.01 (n/a)147.60 (n/a)129.94 (n/a)130.00 (n/a)108.60 (n/a)15.21 (n/a)

mem_copy_1_cores_1_chans_2048_tile_2048_False0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.10 (+31.19%)0.09 (+26.03%)0.10 (+33.11%)0.07 (+25.01%)0.02 (+78.79%)118.10 (-20.04%)96.52 (-19.61%)84.40 (-24.91%)82.40 (-23.77%)18.29 (+9.09%)
0f0da0b — 2026-01-14 16:31:000.08 (n/a)0.07 (n/a)0.07 (n/a)0.06 (n/a)0.01 (n/a)147.70 (n/a)120.06 (n/a)112.40 (n/a)108.10 (n/a)16.76 (n/a)

mem_copy_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
003b49d — 2025-11-14 19:30:540.06 (-15.32%)0.05 (-17.09%)0.05 (-23.81%)0.04 (-15.54%)0.01 (-18.18%)193.00 (+18.40%)160.80 (+20.36%)168.00 (+31.25%)124.00 (+18.10%)22.92 (+11.39%)
0999a9a — 2025-11-13 19:24:080.07 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)163.00 (n/a)133.60 (n/a)128.00 (n/a)105.00 (n/a)20.58 (n/a)

mem_copy_2_cols_2_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
003b49d — 2025-11-14 19:30:540.06 (-23.02%)0.05 (-6.98%)0.05 (-2.13%)0.04 (-15.31%)0.01 (-33.29%)196.00 (+18.07%)147.00 (+6.52%)141.00 (+2.17%)126.00 (+29.90%)25.27 (+8.13%)
0999a9a — 2025-11-13 19:24:080.08 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)166.00 (n/a)138.00 (n/a)138.00 (n/a)97.00 (n/a)23.37 (n/a)

mem_copy_2_cores_1_chans_2048_tile_1024_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.10 (+31.37%)0.08 (+29.12%)0.08 (+29.06%)0.07 (+41.80%)0.01 (+0.23%)109.90 (-29.46%)98.30 (-23.38%)100.80 (-22.52%)79.40 (-23.87%)10.54 (-46.17%)
0a6c11c — 2025-12-04 00:39:100.08 (n/a)0.07 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)155.80 (n/a)128.30 (n/a)130.10 (n/a)104.30 (n/a)19.58 (n/a)

mem_copy_2_cores_1_chans_2048_tile_1024_False0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.12 (+103.55%)0.09 (+81.99%)0.09 (+69.86%)0.07 (+83.95%)0.02 (+187.42%)117.10 (-45.66%)92.20 (-43.53%)86.90 (-41.12%)70.40 (-50.84%)23.32 (-23.43%)
0f0da0b — 2026-01-14 16:31:000.06 (n/a)0.05 (n/a)0.06 (n/a)0.04 (n/a)0.01 (n/a)215.50 (n/a)163.26 (n/a)147.60 (n/a)143.20 (n/a)30.45 (n/a)

mem_copy_2_cores_2_chans_2048_tile_1024_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.10 (+25.02%)0.07 (+10.01%)0.07 (+7.53%)0.05 (-1.99%)0.01 (+77.08%)150.10 (+2.04%)114.60 (-7.22%)111.70 (-6.99%)83.90 (-20.02%)21.23 (+43.54%)
0a6c11c — 2025-12-04 00:39:100.08 (n/a)0.07 (n/a)0.07 (n/a)0.06 (n/a)0.01 (n/a)147.10 (n/a)123.52 (n/a)120.10 (n/a)104.90 (n/a)14.79 (n/a)

mem_copy_2_cores_2_chans_2048_tile_1024_False0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.10 (+35.25%)0.08 (+37.83%)0.09 (+48.91%)0.06 (+9.50%)0.02 (+125.91%)140.20 (-8.66%)101.66 (-25.39%)92.10 (-32.82%)82.90 (-26.11%)24.23 (+51.17%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)153.50 (n/a)136.26 (n/a)137.10 (n/a)112.20 (n/a)16.03 (n/a)

mem_copy_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
003b49d — 2025-11-14 19:30:540.07 (+11.97%)0.05 (+3.09%)0.05 (+2.10%)0.03 (-11.35%)0.01 (+41.65%)229.00 (+12.81%)156.00 (-0.51%)143.00 (-2.05%)117.00 (-10.69%)38.21 (+48.71%)
0999a9a — 2025-11-13 19:24:080.06 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)203.00 (n/a)156.80 (n/a)146.00 (n/a)131.00 (n/a)25.69 (n/a)

mem_copy_4_cols_2_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
003b49d — 2025-11-14 19:30:540.06 (+9.02%)0.05 (+5.59%)0.05 (+0.00%)0.04 (+3.63%)0.01 (+25.48%)193.00 (-3.50%)147.40 (-4.66%)147.00 (+0.00%)122.00 (-8.27%)25.24 (+7.48%)
0999a9a — 2025-11-13 19:24:080.06 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)200.00 (n/a)154.60 (n/a)147.00 (n/a)133.00 (n/a)23.48 (n/a)

mem_copy_4_cores_1_chans_2048_tile_512_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.09 (-3.98%)0.07 (+4.00%)0.07 (+7.04%)0.06 (+3.67%)0.01 (-15.76%)129.70 (-3.57%)113.30 (-4.50%)116.90 (-6.55%)90.40 (+4.15%)15.04 (-13.80%)
0a6c11c — 2025-12-04 00:39:100.09 (n/a)0.07 (n/a)0.07 (n/a)0.06 (n/a)0.01 (n/a)134.50 (n/a)118.64 (n/a)125.10 (n/a)86.80 (n/a)17.45 (n/a)

mem_copy_4_cores_1_chans_2048_tile_512_False0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.12 (+51.68%)0.10 (+55.58%)0.11 (+85.86%)0.06 (+8.92%)0.03 (+163.31%)128.30 (-8.23%)84.56 (-32.56%)71.60 (-46.17%)65.70 (-34.04%)26.42 (+57.86%)
0f0da0b — 2026-01-14 16:31:000.08 (n/a)0.07 (n/a)0.06 (n/a)0.06 (n/a)0.01 (n/a)139.80 (n/a)125.38 (n/a)133.00 (n/a)99.60 (n/a)16.73 (n/a)

mem_copy_4_cores_2_chans_2048_tile_512_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.08 (+9.57%)0.07 (+16.85%)0.07 (+19.58%)0.06 (+24.52%)0.01 (+1.05%)134.50 (-19.70%)116.64 (-14.92%)116.80 (-16.39%)98.00 (-8.75%)15.71 (-25.13%)
0a6c11c — 2025-12-04 00:39:100.08 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)167.50 (n/a)137.10 (n/a)139.70 (n/a)107.40 (n/a)20.99 (n/a)

mem_copy_4_cores_2_chans_2048_tile_512_False0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.10 (+21.52%)0.09 (+51.62%)0.10 (+79.69%)0.06 (+23.77%)0.02 (+25.48%)133.20 (-19.17%)92.98 (-33.88%)83.40 (-44.33%)82.10 (-17.74%)22.49 (-15.24%)
0f0da0b — 2026-01-14 16:31:000.08 (n/a)0.06 (n/a)0.05 (n/a)0.05 (n/a)0.01 (n/a)164.80 (n/a)140.62 (n/a)149.80 (n/a)99.80 (n/a)26.53 (n/a)

mem_copy_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
003b49d — 2025-11-14 19:30:540.05 (-13.48%)0.05 (-17.79%)0.05 (-15.33%)0.03 (-25.09%)0.01 (+6.67%)267.00 (+33.50%)176.80 (+24.51%)150.00 (+18.11%)141.00 (+15.57%)47.41 (+61.41%)
0999a9a — 2025-11-13 19:24:080.06 (n/a)0.06 (n/a)0.06 (n/a)0.04 (n/a)0.01 (n/a)200.00 (n/a)142.00 (n/a)127.00 (n/a)122.00 (n/a)29.37 (n/a)

mem_copy_8_cols_2_channels_2048_tile_128

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
003b49d — 2025-11-14 19:30:540.05 (+13.38%)0.05 (+23.85%)0.05 (+37.91%)0.05 (+30.72%)0.00 (-56.85%)166.00 (-23.50%)153.40 (-20.60%)153.00 (-27.49%)142.00 (-11.80%)7.63 (-70.60%)
0999a9a — 2025-11-13 19:24:080.05 (n/a)0.04 (n/a)0.04 (n/a)0.04 (n/a)0.01 (n/a)217.00 (n/a)193.20 (n/a)211.00 (n/a)161.00 (n/a)25.96 (n/a)

mem_copy_8_cores_1_chans_2048_tile_256_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.09 (+16.31%)0.08 (+15.34%)0.08 (+14.60%)0.07 (+19.30%)0.01 (-5.41%)117.30 (-16.15%)107.80 (-13.61%)108.50 (-12.78%)94.10 (-13.99%)8.29 (-32.01%)
0a6c11c — 2025-12-04 00:39:100.07 (n/a)0.07 (n/a)0.07 (n/a)0.06 (n/a)0.01 (n/a)139.90 (n/a)124.78 (n/a)124.40 (n/a)109.40 (n/a)12.20 (n/a)

mem_copy_8_cores_1_chans_2048_tile_256_False0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.09 (+46.69%)0.08 (+57.18%)0.09 (+64.02%)0.06 (+94.45%)0.01 (+6.58%)131.30 (-48.59%)99.78 (-38.77%)88.70 (-39.00%)86.90 (-31.84%)18.90 (-64.18%)
0f0da0b — 2026-01-14 16:31:000.06 (n/a)0.05 (n/a)0.06 (n/a)0.03 (n/a)0.01 (n/a)255.40 (n/a)162.96 (n/a)145.40 (n/a)127.50 (n/a)52.78 (n/a)

mem_copy_8_cores_2_chans_2048_tile_256_False

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.09 (+12.61%)0.07 (-2.24%)0.07 (-8.85%)0.06 (+5.13%)0.01 (+35.50%)136.50 (-4.88%)120.86 (+2.96%)125.10 (+9.64%)91.60 (-11.15%)15.72 (+10.46%)
0a6c11c — 2025-12-04 00:39:100.08 (n/a)0.07 (n/a)0.07 (n/a)0.06 (n/a)0.01 (n/a)143.50 (n/a)117.38 (n/a)114.10 (n/a)103.10 (n/a)14.23 (n/a)

mem_copy_8_cores_2_chans_2048_tile_256_False0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.10 (+63.97%)0.09 (+58.77%)0.09 (+53.52%)0.06 (+50.07%)0.02 (+70.34%)144.40 (-33.36%)97.70 (-36.49%)88.10 (-34.89%)80.00 (-39.02%)26.35 (-27.55%)
0f0da0b — 2026-01-14 16:31:000.06 (n/a)0.06 (n/a)0.06 (n/a)0.04 (n/a)0.01 (n/a)216.70 (n/a)153.84 (n/a)135.30 (n/a)131.20 (n/a)36.36 (n/a)

mha

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.18 (+0.16%)0.18 (+0.19%)0.18 (+0.12%)0.18 (+0.32%)0.00 (-52.24%)47253.90 (-0.32%)47213.70 (-0.19%)47210.90 (-0.12%)47167.70 (-0.16%)28.71 (-52.47%)
0a6c11c — 2025-12-04 00:39:100.18 (n/a)0.18 (n/a)0.18 (n/a)0.18 (n/a)0.00 (n/a)47405.00 (n/a)47301.72 (n/a)47265.90 (n/a)47241.80 (n/a)60.40 (n/a)

mha0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.18 (-0.08%)0.18 (+0.22%)0.18 (+0.32%)0.18 (+0.35%)0.00 (-42.58%)47367.50 (-0.34%)47232.68 (-0.22%)47232.00 (-0.32%)47073.00 (+0.08%)105.45 (-42.71%)
0f0da0b — 2026-01-14 16:31:000.18 (n/a)0.18 (n/a)0.18 (n/a)0.18 (n/a)0.00 (n/a)47531.20 (n/a)47335.18 (n/a)47384.30 (n/a)47037.20 (n/a)184.08 (n/a)

relu_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.11 (+69.81%)0.09 (+59.10%)0.10 (+75.49%)0.05 (+26.24%)0.03 (+206.46%)152.90 (-20.82%)101.62 (-32.81%)82.20 (-43.00%)73.00 (-41.13%)35.75 (+36.67%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.06 (n/a)0.06 (n/a)0.04 (n/a)0.01 (n/a)193.10 (n/a)151.24 (n/a)144.20 (n/a)124.00 (n/a)26.16 (n/a)

relu_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.10 (+60.88%)0.08 (+49.18%)0.09 (+55.13%)0.05 (+8.09%)0.02 (+316.89%)150.60 (-7.44%)101.26 (-30.06%)91.30 (-35.57%)83.00 (-37.87%)28.02 (+146.84%)
0f0da0b — 2026-01-14 16:31:000.06 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.00 (n/a)162.70 (n/a)144.78 (n/a)141.70 (n/a)133.60 (n/a)11.35 (n/a)

relu_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.10 (+36.44%)0.09 (+44.22%)0.09 (+49.00%)0.06 (+25.29%)0.02 (+65.69%)135.80 (-20.21%)96.20 (-29.75%)86.30 (-32.84%)83.10 (-26.72%)22.46 (-2.41%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)170.20 (n/a)136.94 (n/a)128.50 (n/a)113.40 (n/a)23.02 (n/a)

relu_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.11 (+58.81%)0.09 (+71.41%)0.10 (+74.46%)0.05 (+55.08%)0.02 (+79.07%)154.40 (-35.51%)93.32 (-40.67%)78.40 (-42.65%)75.80 (-37.04%)34.24 (-28.56%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.06 (n/a)0.06 (n/a)0.03 (n/a)0.01 (n/a)239.40 (n/a)157.28 (n/a)136.70 (n/a)120.40 (n/a)47.93 (n/a)

rms_norm_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.09 (+27.92%)0.08 (+35.45%)0.09 (+45.35%)0.06 (+20.87%)0.01 (+35.02%)131.50 (-17.30%)100.70 (-25.92%)92.40 (-31.20%)90.90 (-21.77%)17.35 (-10.74%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)159.00 (n/a)135.94 (n/a)134.30 (n/a)116.20 (n/a)19.43 (n/a)

rms_norm_1_cols_2_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.12 (+67.22%)0.10 (+57.83%)0.11 (+74.39%)0.06 (-3.10%)0.03 (+484.78%)145.20 (+3.20%)86.54 (-31.66%)72.10 (-42.64%)70.50 (-40.25%)32.80 (+266.62%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.06 (n/a)0.07 (n/a)0.06 (n/a)0.00 (n/a)140.70 (n/a)126.64 (n/a)125.70 (n/a)118.00 (n/a)8.95 (n/a)

rms_norm_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.09 (+26.55%)0.09 (+40.18%)0.09 (+47.01%)0.06 (+19.23%)0.01 (+54.24%)129.60 (-16.12%)98.22 (-28.12%)90.90 (-31.96%)89.80 (-20.95%)17.55 (+0.58%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)154.50 (n/a)136.64 (n/a)133.60 (n/a)113.60 (n/a)17.45 (n/a)

rms_norm_2_cols_2_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.13 (+67.61%)0.11 (+72.13%)0.11 (+66.55%)0.09 (+90.95%)0.02 (+33.97%)93.50 (-47.62%)74.32 (-42.73%)73.10 (-39.98%)62.40 (-40.34%)11.60 (-59.00%)
0f0da0b — 2026-01-14 16:31:000.08 (n/a)0.07 (n/a)0.07 (n/a)0.05 (n/a)0.01 (n/a)178.50 (n/a)129.78 (n/a)121.80 (n/a)104.60 (n/a)28.30 (n/a)

rms_norm_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.10 (+60.49%)0.09 (+76.01%)0.09 (+72.18%)0.06 (+64.97%)0.01 (+31.34%)130.70 (-39.41%)97.48 (-43.89%)90.20 (-41.92%)86.00 (-37.68%)18.66 (-50.69%)
0f0da0b — 2026-01-14 16:31:000.06 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)215.70 (n/a)173.74 (n/a)155.30 (n/a)138.00 (n/a)37.85 (n/a)

rms_norm_4_cols_2_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.11 (+60.28%)0.10 (+80.30%)0.11 (+105.91%)0.07 (+52.87%)0.02 (+90.94%)125.50 (-34.60%)83.90 (-43.66%)73.50 (-51.42%)72.90 (-37.64%)23.26 (-20.88%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.06 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)191.90 (n/a)148.92 (n/a)151.30 (n/a)116.90 (n/a)29.40 (n/a)

rms_norm_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.09 (+27.54%)0.08 (+37.27%)0.09 (+53.73%)0.05 (+11.36%)0.01 (+87.66%)154.30 (-10.24%)106.66 (-25.46%)94.80 (-34.98%)94.00 (-21.60%)26.64 (+33.21%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)171.90 (n/a)143.10 (n/a)145.80 (n/a)119.90 (n/a)20.00 (n/a)

rms_norm_8_cols_2_channels_2048_tile_128

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.09 (+38.93%)0.08 (+46.69%)0.09 (+73.48%)0.05 (+5.47%)0.02 (+140.19%)167.70 (-5.15%)109.36 (-28.83%)91.60 (-42.35%)88.90 (-28.02%)33.52 (+68.39%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.05 (n/a)0.05 (n/a)0.05 (n/a)0.01 (n/a)176.80 (n/a)153.66 (n/a)158.90 (n/a)123.50 (n/a)19.91 (n/a)

rope_1_cols_2_channels_4096_tile_4096_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1b2d230 — 2026-01-13 22:38:030.19 (-21.66%)0.17 (-8.53%)0.18 (+2.09%)0.14 (-6.71%)0.02 (-38.07%)174.10 (+7.20%)150.06 (+8.11%)139.60 (-2.04%)129.80 (+27.63%)20.36 (-10.65%)
d48746f — 2026-01-12 22:17:100.24 (n/a)0.18 (n/a)0.17 (n/a)0.15 (n/a)0.04 (n/a)162.40 (n/a)138.80 (n/a)142.50 (n/a)101.70 (n/a)22.79 (n/a)

rope_1c_32rows_512cols_32arows_0m

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:241.10 (+30.53%)0.99 (+42.49%)1.05 (+62.73%)0.73 (+21.76%)0.15 (+50.52%)133.80 (-17.86%)101.54 (-29.34%)93.30 (-38.58%)89.10 (-23.39%)18.60 (-3.62%)
0f0da0b — 2026-01-14 16:31:000.85 (n/a)0.69 (n/a)0.65 (n/a)0.60 (n/a)0.10 (n/a)162.90 (n/a)143.70 (n/a)151.90 (n/a)116.30 (n/a)19.30 (n/a)

rope_1c_32rows_512cols_8arows_0m

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.56 (-21.50%)0.52 (-8.95%)0.53 (-6.60%)0.48 (+7.05%)0.03 (-70.73%)153.70 (-6.57%)140.84 (+7.27%)140.10 (+7.03%)132.50 (+27.40%)8.40 (-65.00%)
0f0da0b — 2026-01-14 16:31:000.71 (n/a)0.58 (n/a)0.56 (n/a)0.45 (n/a)0.10 (n/a)164.50 (n/a)131.30 (n/a)130.90 (n/a)104.00 (n/a)24.00 (n/a)

rope_2_cols_2_channels_4096_tile_2048_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1b2d230 — 2026-01-13 22:38:030.21 (-0.56%)0.17 (-10.62%)0.16 (-13.58%)0.13 (-13.33%)0.03 (+17.05%)183.70 (+15.39%)151.76 (+12.77%)155.20 (+15.73%)117.30 (+0.60%)24.24 (+35.73%)
d48746f — 2026-01-12 22:17:100.21 (n/a)0.19 (n/a)0.18 (n/a)0.15 (n/a)0.02 (n/a)159.20 (n/a)134.58 (n/a)134.10 (n/a)116.60 (n/a)17.86 (n/a)

rope_2c_32rows_512cols_32arows_0m

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:241.25 (+43.66%)0.97 (+36.64%)0.91 (+36.32%)0.63 (+2.79%)0.25 (+150.75%)157.00 (-2.67%)108.10 (-23.42%)108.30 (-26.68%)78.70 (-30.42%)31.09 (+71.34%)
0f0da0b — 2026-01-14 16:31:000.87 (n/a)0.71 (n/a)0.67 (n/a)0.61 (n/a)0.10 (n/a)161.30 (n/a)141.16 (n/a)147.70 (n/a)113.10 (n/a)18.15 (n/a)

rope_2c_32rows_512cols_8arows_0m

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.63 (-10.72%)0.57 (+15.63%)0.57 (+20.61%)0.49 (+35.05%)0.05 (-60.20%)150.60 (-25.96%)130.58 (-17.08%)130.40 (-17.05%)116.80 (+12.09%)12.57 (-65.56%)
0f0da0b — 2026-01-14 16:31:000.71 (n/a)0.49 (n/a)0.47 (n/a)0.36 (n/a)0.13 (n/a)203.40 (n/a)157.48 (n/a)157.20 (n/a)104.20 (n/a)36.49 (n/a)

rope_4_cols_2_channels_4096_tile_1024_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1b2d230 — 2026-01-13 22:38:030.22 (+18.92%)0.17 (+6.24%)0.17 (+5.57%)0.14 (-0.50%)0.03 (+90.77%)171.10 (+0.47%)147.40 (-4.39%)148.10 (-5.31%)110.10 (-15.89%)23.59 (+60.67%)
d48746f — 2026-01-12 22:17:100.19 (n/a)0.16 (n/a)0.16 (n/a)0.14 (n/a)0.02 (n/a)170.30 (n/a)154.16 (n/a)156.40 (n/a)130.90 (n/a)14.69 (n/a)

rope_8_cols_2_channels_4096_tile_512_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1b2d230 — 2026-01-13 22:38:030.17 (-1.30%)0.15 (+6.63%)0.15 (-8.78%)0.14 (+39.68%)0.01 (-67.35%)174.90 (-28.41%)163.42 (-10.36%)167.40 (+9.63%)146.50 (+1.31%)11.19 (-76.17%)
d48746f — 2026-01-12 22:17:100.17 (n/a)0.14 (n/a)0.16 (n/a)0.10 (n/a)0.03 (n/a)244.30 (n/a)182.30 (n/a)152.70 (n/a)144.60 (n/a)46.98 (n/a)

rope_8c_32rows_512cols_32arows_0m

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:241.34 (+62.86%)0.89 (+47.92%)0.85 (+53.70%)0.60 (+49.43%)0.27 (+64.12%)163.50 (-33.07%)117.38 (-32.11%)115.50 (-34.93%)73.60 (-38.56%)32.08 (-33.78%)
0f0da0b — 2026-01-14 16:31:000.82 (n/a)0.60 (n/a)0.55 (n/a)0.40 (n/a)0.16 (n/a)244.30 (n/a)172.90 (n/a)177.50 (n/a)119.80 (n/a)48.44 (n/a)

rope_8c_32rows_512cols_8arows_0m

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.54 (-10.87%)0.52 (+6.74%)0.53 (+12.23%)0.48 (+19.79%)0.03 (-67.26%)152.90 (-16.49%)143.32 (-8.02%)139.70 (-10.91%)135.50 (+12.17%)7.46 (-68.88%)
0f0da0b — 2026-01-14 16:31:000.61 (n/a)0.48 (n/a)0.47 (n/a)0.40 (n/a)0.08 (n/a)183.10 (n/a)155.82 (n/a)156.80 (n/a)120.80 (n/a)23.99 (n/a)

rows_32-cols_512-angle_rows_32-aie_columns_1-method_type_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3cb43db — 2026-02-10 01:03:390.35 (n/a)0.27 (n/a)0.27 (n/a)0.20 (n/a)0.05 (n/a)479.90 (n/a)370.30 (n/a)370.70 (n/a)284.00 (n/a)71.58 (n/a)

rows_32-cols_512-angle_rows_32-aie_columns_2-method_type_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3cb43db — 2026-02-10 01:03:390.39 (n/a)0.30 (n/a)0.30 (n/a)0.22 (n/a)0.06 (n/a)440.60 (n/a)333.86 (n/a)327.30 (n/a)252.70 (n/a)71.28 (n/a)

rows_32-cols_512-angle_rows_32-aie_columns_8-method_type_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3cb43db — 2026-02-10 01:03:390.36 (n/a)0.26 (n/a)0.27 (n/a)0.19 (n/a)0.07 (n/a)520.50 (n/a)393.58 (n/a)365.00 (n/a)269.40 (n/a)97.45 (n/a)

rows_32-cols_512-angle_rows_8-aie_columns_1-method_type_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3cb43db — 2026-02-10 01:03:390.25 (n/a)0.23 (n/a)0.23 (n/a)0.21 (n/a)0.02 (n/a)351.40 (n/a)318.20 (n/a)315.90 (n/a)294.70 (n/a)21.72 (n/a)

rows_32-cols_512-angle_rows_8-aie_columns_2-method_type_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3cb43db — 2026-02-10 01:03:390.27 (n/a)0.24 (n/a)0.25 (n/a)0.17 (n/a)0.04 (n/a)427.00 (n/a)321.02 (n/a)299.50 (n/a)275.40 (n/a)60.30 (n/a)

rows_32-cols_512-angle_rows_8-aie_columns_8-method_type_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3cb43db — 2026-02-10 01:03:390.26 (n/a)0.17 (n/a)0.16 (n/a)0.13 (n/a)0.05 (n/a)589.60 (n/a)460.86 (n/a)452.50 (n/a)285.90 (n/a)126.50 (n/a)

seq_len_16384-dim_64-num_heads_1-num_pipelines_8

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
3cb43db — 2026-02-10 01:03:390.17 (n/a)0.17 (n/a)0.17 (n/a)0.17 (n/a)0.00 (n/a)48513.70 (n/a)48327.64 (n/a)48352.20 (n/a)48136.70 (n/a)145.36 (n/a)

seq_len_256-embedding_dim_2048-hidden_dim_2048-prio_accuracy_False

No metrics available.

sigmoid_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.12 (+53.53%)0.10 (+54.49%)0.12 (+106.52%)0.05 (-0.16%)0.03 (+144.85%)153.20 (+0.13%)90.92 (-30.39%)68.60 (-51.55%)68.30 (-34.89%)36.80 (+58.03%)
0f0da0b — 2026-01-14 16:31:000.08 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)153.00 (n/a)130.62 (n/a)141.60 (n/a)104.90 (n/a)23.29 (n/a)

sigmoid_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.12 (+73.69%)0.10 (+60.60%)0.12 (+98.01%)0.06 (+2.86%)0.03 (+394.96%)140.80 (-2.83%)89.84 (-32.71%)68.90 (-49.52%)68.10 (-42.43%)32.04 (+163.16%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.06 (n/a)0.06 (n/a)0.06 (n/a)0.01 (n/a)144.90 (n/a)133.52 (n/a)136.50 (n/a)118.30 (n/a)12.18 (n/a)

sigmoid_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.10 (+30.15%)0.08 (+37.20%)0.09 (+57.41%)0.05 (+2.94%)0.02 (+110.00%)169.40 (-2.87%)106.58 (-23.47%)87.40 (-36.48%)84.00 (-23.15%)36.37 (+56.68%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)174.40 (n/a)139.26 (n/a)137.60 (n/a)109.30 (n/a)23.21 (n/a)

sigmoid_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.12 (+109.66%)0.10 (+84.35%)0.12 (+117.89%)0.05 (+3.89%)0.03 (+649.09%)157.30 (-3.73%)89.74 (-39.88%)68.70 (-54.11%)65.90 (-52.35%)38.85 (+251.55%)
0f0da0b — 2026-01-14 16:31:000.06 (n/a)0.06 (n/a)0.05 (n/a)0.05 (n/a)0.00 (n/a)163.40 (n/a)149.28 (n/a)149.70 (n/a)138.30 (n/a)11.05 (n/a)

silu_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.10 (+44.80%)0.08 (+33.22%)0.10 (+66.93%)0.06 (-1.08%)0.02 (+301.40%)143.30 (+1.06%)104.88 (-20.01%)82.20 (-40.13%)80.60 (-30.93%)32.24 (+177.47%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.06 (n/a)0.06 (n/a)0.06 (n/a)0.01 (n/a)141.80 (n/a)131.12 (n/a)137.30 (n/a)116.70 (n/a)11.62 (n/a)

silu_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.10 (+34.12%)0.08 (+47.27%)0.10 (+69.15%)0.06 (+39.46%)0.02 (+73.50%)138.90 (-28.29%)103.10 (-30.67%)83.20 (-40.91%)81.60 (-25.41%)28.30 (-8.88%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.06 (n/a)0.06 (n/a)0.04 (n/a)0.01 (n/a)193.70 (n/a)148.70 (n/a)140.80 (n/a)109.40 (n/a)31.06 (n/a)

silu_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.12 (+42.12%)0.11 (+61.07%)0.12 (+92.74%)0.06 (+8.38%)0.03 (+104.00%)137.40 (-7.72%)83.36 (-34.98%)70.00 (-48.07%)68.10 (-29.65%)30.27 (+32.68%)
0f0da0b — 2026-01-14 16:31:000.08 (n/a)0.07 (n/a)0.06 (n/a)0.06 (n/a)0.01 (n/a)148.90 (n/a)128.20 (n/a)134.80 (n/a)96.80 (n/a)22.82 (n/a)

silu_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.12 (+35.33%)0.09 (+59.58%)0.09 (+78.47%)0.07 (+60.20%)0.02 (-1.76%)109.80 (-37.58%)89.20 (-38.82%)87.90 (-43.98%)71.10 (-26.09%)14.50 (-53.17%)
0f0da0b — 2026-01-14 16:31:000.09 (n/a)0.06 (n/a)0.05 (n/a)0.05 (n/a)0.02 (n/a)175.90 (n/a)145.80 (n/a)156.90 (n/a)96.20 (n/a)30.97 (n/a)

softmax_1_cols_2_channels_4096_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.21 (+20.19%)0.16 (+13.84%)0.16 (+8.19%)0.12 (+26.61%)0.04 (+17.18%)135.50 (-20.99%)104.68 (-12.65%)102.70 (-7.56%)76.50 (-16.85%)22.55 (-26.21%)
0f0da0b — 2026-01-14 16:31:000.18 (n/a)0.14 (n/a)0.15 (n/a)0.10 (n/a)0.03 (n/a)171.50 (n/a)119.84 (n/a)111.10 (n/a)92.00 (n/a)30.56 (n/a)

softmax_2_cols_2_channels_4096_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.22 (+40.26%)0.18 (+43.33%)0.20 (+64.08%)0.13 (+24.34%)0.04 (+68.98%)127.80 (-19.62%)94.74 (-28.95%)83.40 (-39.08%)73.90 (-28.67%)24.46 (-5.31%)
0f0da0b — 2026-01-14 16:31:000.16 (n/a)0.13 (n/a)0.12 (n/a)0.10 (n/a)0.03 (n/a)159.00 (n/a)133.34 (n/a)136.90 (n/a)103.60 (n/a)25.83 (n/a)

softmax_2_cols_2_channels_4096_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.24 (+85.21%)0.17 (+69.97%)0.15 (+54.16%)0.11 (+37.21%)0.05 (+203.82%)148.10 (-27.12%)102.78 (-37.88%)106.70 (-35.10%)69.60 (-45.96%)31.81 (+15.98%)
0f0da0b — 2026-01-14 16:31:000.13 (n/a)0.10 (n/a)0.10 (n/a)0.08 (n/a)0.02 (n/a)203.20 (n/a)165.46 (n/a)164.40 (n/a)128.80 (n/a)27.43 (n/a)

swiglu

No metrics available.

swiglu_decode_1x2048x2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:490.00 (-15.46%)0.00 (-27.20%)0.00 (-28.17%)0.00 (-33.90%)0.00 (+1834.83%)5187.20 (+51.28%)4720.84 (+38.38%)4750.90 (+39.22%)4015.10 (+18.29%)390.23 (+3298.45%)
0a6c11c — 2025-12-04 00:39:100.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)3428.80 (n/a)3411.62 (n/a)3412.60 (n/a)3394.20 (n/a)11.48 (n/a)

swiglu_decode_1x2048x2048_0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.00 (+28.14%)0.00 (+20.30%)0.00 (+10.00%)0.00 (+11.66%)0.00 (+124.73%)5111.40 (-10.44%)4324.88 (-14.91%)4736.10 (-9.09%)3410.70 (-21.96%)843.97 (+53.51%)
0f0da0b — 2026-01-14 16:31:000.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)5707.20 (n/a)5083.00 (n/a)5209.60 (n/a)4370.50 (n/a)549.78 (n/a)

tanh_1_cols_1_channels_2048_tile_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.10 (+42.65%)0.08 (+33.21%)0.07 (+27.20%)0.06 (+13.46%)0.02 (+130.64%)140.80 (-11.89%)109.14 (-23.00%)115.90 (-21.37%)85.60 (-29.89%)23.52 (+37.00%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)159.80 (n/a)141.74 (n/a)147.40 (n/a)122.10 (n/a)17.17 (n/a)

tanh_2_cols_1_channels_2048_tile_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.11 (+41.27%)0.09 (+40.30%)0.09 (+45.94%)0.08 (+28.08%)0.01 (+83.15%)103.30 (-21.98%)88.02 (-28.44%)87.80 (-31.46%)77.80 (-29.21%)9.64 (+2.40%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.07 (n/a)0.06 (n/a)0.06 (n/a)0.01 (n/a)132.40 (n/a)123.00 (n/a)128.10 (n/a)109.90 (n/a)9.41 (n/a)

tanh_4_cols_1_channels_2048_tile_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.10 (+48.49%)0.09 (+56.51%)0.09 (+68.30%)0.07 (+54.99%)0.01 (+44.32%)121.40 (-35.46%)93.40 (-36.24%)86.30 (-40.61%)84.60 (-32.70%)15.80 (-36.89%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.06 (n/a)0.06 (n/a)0.04 (n/a)0.01 (n/a)188.10 (n/a)146.48 (n/a)145.30 (n/a)125.70 (n/a)25.04 (n/a)

tanh_8_cols_1_channels_2048_tile_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.11 (+64.36%)0.09 (+41.14%)0.10 (+56.21%)0.05 (-7.33%)0.03 (+346.69%)163.80 (+7.91%)102.90 (-22.05%)80.40 (-35.94%)71.70 (-39.13%)39.99 (+185.25%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.06 (n/a)0.07 (n/a)0.05 (n/a)0.01 (n/a)151.80 (n/a)132.00 (n/a)125.50 (n/a)117.80 (n/a)14.02 (n/a)

transpose_2048_M_64_N_1_cols_1_channels_64_m_64_n_8_s

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:493.93 (-7.02%)3.68 (+2.48%)3.66 (+1.82%)3.45 (+15.62%)0.15 (-63.64%)151.80 (-13.50%)142.70 (-3.59%)143.40 (-1.78%)133.50 (+7.49%)5.90 (-66.22%)
0a6c11c — 2025-12-04 00:39:104.22 (n/a)3.59 (n/a)3.59 (n/a)2.99 (n/a)0.42 (n/a)175.50 (n/a)148.02 (n/a)146.00 (n/a)124.20 (n/a)17.46 (n/a)

transpose_2048_M_64_N_1_cols_1_channels_64_m_64_n_8_s0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:244.95 (+35.28%)4.34 (+33.32%)4.37 (+25.58%)3.74 (+35.73%)0.50 (+11.62%)140.30 (-26.35%)121.98 (-25.39%)119.90 (-20.33%)106.00 (-26.08%)14.15 (-39.67%)
0f0da0b — 2026-01-14 16:31:003.66 (n/a)3.26 (n/a)3.48 (n/a)2.75 (n/a)0.45 (n/a)190.50 (n/a)163.48 (n/a)150.50 (n/a)143.40 (n/a)23.46 (n/a)

transpose_2048_M_64_N_1_cols_2_channels_64_m_64_n_8_s

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
130b6ea — 2025-12-05 21:15:495.11 (+15.57%)3.85 (-3.32%)3.40 (-18.64%)3.29 (+7.09%)0.71 (+49.14%)159.60 (-6.61%)140.18 (+4.80%)154.20 (+22.97%)102.60 (-13.49%)22.70 (+19.77%)
0a6c11c — 2025-12-04 00:39:104.42 (n/a)3.99 (n/a)4.18 (n/a)3.07 (n/a)0.48 (n/a)170.90 (n/a)133.76 (n/a)125.40 (n/a)118.60 (n/a)18.95 (n/a)

transpose_2048_M_64_N_1_cols_2_channels_64_m_64_n_8_s0

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:246.07 (+44.96%)4.99 (+45.77%)5.09 (+52.05%)4.09 (+35.80%)0.74 (+53.13%)128.10 (-26.34%)106.86 (-31.23%)103.00 (-34.27%)86.30 (-31.02%)15.67 (-22.94%)
0f0da0b — 2026-01-14 16:31:004.19 (n/a)3.42 (n/a)3.35 (n/a)3.01 (n/a)0.48 (n/a)173.90 (n/a)155.38 (n/a)156.70 (n/a)125.10 (n/a)20.34 (n/a)

weighted_rms_norm_1_cols_2_channels_2048_weights_2048

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.11 (+41.27%)0.09 (+39.11%)0.09 (+47.90%)0.05 (+1.24%)0.02 (+132.19%)155.00 (-1.21%)99.92 (-24.84%)89.60 (-32.43%)76.50 (-29.17%)31.36 (+74.29%)
0f0da0b — 2026-01-14 16:31:000.08 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)156.90 (n/a)132.94 (n/a)132.60 (n/a)108.00 (n/a)17.99 (n/a)

weighted_rms_norm_2_cols_2_channels_2048_weights_1024

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.10 (+48.28%)0.09 (+48.11%)0.09 (+60.52%)0.05 (+2.82%)0.02 (+134.54%)166.70 (-2.74%)102.72 (-28.96%)88.50 (-37.68%)83.10 (-32.55%)35.87 (+61.58%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)171.40 (n/a)144.60 (n/a)142.00 (n/a)123.20 (n/a)22.20 (n/a)

weighted_rms_norm_4_cols_2_channels_2048_weights_512

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.09 (+21.03%)0.09 (+31.88%)0.09 (+37.19%)0.07 (+25.91%)0.01 (+25.73%)119.00 (-20.61%)96.48 (-24.14%)89.90 (-27.09%)89.30 (-17.31%)12.81 (-17.86%)
0f0da0b — 2026-01-14 16:31:000.08 (n/a)0.07 (n/a)0.07 (n/a)0.05 (n/a)0.01 (n/a)149.90 (n/a)127.18 (n/a)123.30 (n/a)108.00 (n/a)15.59 (n/a)

weighted_rms_norm_8_cols_2_channels_2048_weights_256

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
1531781 — 2026-01-23 18:49:240.11 (+67.16%)0.10 (+80.58%)0.10 (+80.44%)0.06 (+74.63%)0.02 (+72.16%)135.80 (-42.72%)89.92 (-44.57%)80.00 (-44.60%)74.30 (-40.23%)25.76 (-41.27%)
0f0da0b — 2026-01-14 16:31:000.07 (n/a)0.05 (n/a)0.06 (n/a)0.03 (n/a)0.01 (n/a)237.10 (n/a)162.22 (n/a)144.40 (n/a)124.30 (n/a)43.86 (n/a)

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@andrej andrej Awaiting requested review from andrej andrej will be requested when the pull request is marked ready for review andrej is a code owner

@jgmelber jgmelber Awaiting requested review from jgmelber jgmelber will be requested when the pull request is marked ready for review jgmelber is a code owner

At least 1 approving review is required to merge this pull request.

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

2 participants

@hunhoffe @andrej