fix(deps): update module github.com/ollama/ollama to v0.21.2

[renovate]

ℹ️ Note

This PR body was truncated due to platform limits.

This PR contains the following updates:

Package	Change	Age	Confidence
github.com/ollama/ollama	v0.3.6 → v0.21.2

---

Release Notes

ollama/ollama (github.com/ollama/ollama)

v0.21.2

Compare Source

What's Changed

• Improved reliability of the OpenClaw onboarding flow in ollama launch
• Recommended models in ollama launch now appear in a fixed, canonical order
• OpenClaw integration now bundles Ollama's web search plugin in OpenClaw

New Contributors

• @​madflow made their first contribution in #​15733

Full Changelog: ollama/ollama@v0.21.1...v0.21.2

v0.21.1

Compare Source

What's Changed

Kimi CLI

You can now install and run the Kimi CLI through Ollama.

ollama launch kimi --model kimi-k2.6:cloud

Kimi CLI with Kimi K2.6 excels at long horizon agentic execution tasks through a multi-agent system.

• MLX runner adds logprobs support for compatible models
• Faster MLX sampling with fused top-P and top-K in a single sort pass, plus repeat penalties applied in the sampler
• Improved MLX prompt tokenization by moving tokenization into request handler goroutines
• Better MLX thread safety for array management
• GLM4 MoE Lite performance improvement with a fused sigmoid router head
• Fixed model picker showing stale model after switching chats in the macOS app
• Fixed structured outputs for Gemma 4 when think=false

Full Changelog: ollama/ollama@v0.21.0...v0.21.1

v0.21.0

Compare Source

Hermes Agent

ollama launch hermes

Hermes learns with you, automatically creating skills to better serve your workflows. Great for research and engineering tasks.

What's Changed

• Gemma 4 on MLX. Added support for running Gemma 4 via MLX on Apple Silicon, including a text-only MLX runtime for the model. The MLX backend also picked up mixed-precision quantization, better capability detection, and a batch of new op wrappers (Conv2d, Pad, activations, trig, masked SDPA, and RoPE-with-freqs).
• Hermes and GitHub Copilot CLI in ollama launch. Added both integrations, which can now be configured in one command alongside the rest of the supported coding agents.
• OpenCode moved to inline config. ollama launch opencode now writes its config inline rather than to a separate file, matching how other integrations are handled.
• ollama launch no longer rewrites config when nothing changed. Pressing → on a configured multi-model integration, or passing --model with the current primary, used to trigger a confirmation prompt and rewrite both the editor's config file and config.json. Now it's a no-op when the resolved model list matches what's already saved.
• Fixed ollama launch openclaw --yes so it correctly skips the channels configuration step, so non-interactive setups complete cleanly.
• Restored the Gemma 4 nothink renderer with the e2b-style prompt.
• Fixed the Gemma 4 compiler error that was breaking Metal builds.
• Fixed macOS cross-compiles so they no longer trigger generate, which was breaking cmake builds on some Xcode versions.
• Quieted cgo builds by suppressing deprecated warnings during go build.

Full Changelog: ollama/ollama@v0.20.7...v0.21.0

v0.20.7

Compare Source

What's Changed

• Fix quality of gemma:e2b and gemma:e4b when thinking is disabled
• ROCm: Update to ROCm 7.2.1 on Linux by @​saman-amd in #​15483

Full Changelog: ollama/ollama@v0.20.6...v0.20.7

v0.20.6

Compare Source

What's Changed

• Gemma 4 tool calling ability is improved and updated to use Google's latest post-launch fixes
• Parallel tool calling improved for streaming responses
• Hermes agent Ollama integration guide is now available
• Ollama app is updated to fix image attachment errors

New Contributors

@​matteocelani made their first contribution in #​15272

Full Changelog: ollama/ollama@v0.20.5...v0.20.6

v0.20.5

Compare Source

OpenClaw channel setup with ollama launch

What's Changed

• OpenClaw channel setup: connect WhatsApp, Telegram, Discord, and other messaging channels through ollama launch openclaw
• Enable flash attention for Gemma 4 on compatible GPUs
• ollama launch opencode now detects curl-based OpenCode installs at ~/.opencode/bin
• Fix /save command for models imported from safetensors

New Contributors

• @​sjhddh made their first contributihttps://github.com/ollama/ollama/pull/15424/15424

Full Changelog: ollama/ollama@v0.20.4...v0.20.5

v0.20.4

Compare Source

What's Changed

• mlx: Improve M5 performance with NAX
• gemma4: enable flash attention

Full Changelog: ollama/ollama@v0.20.3...v0.20.4

v0.20.3

Compare Source

What's Changed

• Gemma 4 Tool Calling improvements
• Added latest models to Ollama App
• OpenClaw fixes for launching TUI

Full Changelog: ollama/ollama@v0.20.2...v0.20.3

v0.20.2

Compare Source

What's Changed

• app: default app home view to new chat instead of launch by @​jmorganca in #​15312

Full Changelog: ollama/ollama@v0.20.1...v0.20.2

v0.20.1

Compare Source

What's Changed

• bench: add prompt calibration, context size flag, and NumCtx reporting by @​dhiltgen in #​15158
• model/parsers: fix gemma4 arg parsing when quoted strings contain " by @​drifkin in #​15254
• ggml: skip cublasGemmBatchedEx during graph reservation by @​jessegross in #​15301
• gemma4: enable flash attention by @​dhiltgen in #​15296
• ggml: fix ROCm build for cublasGemmBatchedEx reserve wrapper by @​jessegross in #​15305
• model/parsers: rework gemma4 tool call handling by @​drifkin in #​15306

Full Changelog: ollama/ollama@v0.20.0...v0.20.1

v0.20.0

Compare Source

Gemma 4

Effective 2B (E2B)

ollama run gemma4:e2b

Effective 4B (E4B)

ollama run gemma4:e4b

26B (Mixture of Experts model with 4B active parameters)

ollama run gemma4:26b

31B (Dense)

ollama run gemma4:31b

What's Changed

• docs: update pi docs by @​ParthSareen in #​15152
• mlx: respect tokenizer add_bos_token setting in pipeline by @​dhiltgen in #​15185
• tokenizer: add SentencePiece-style BPE support by @​dhiltgen in #​15162

Full Changelog: ollama/ollama@v0.19.0...v0.20.0-rc0

v0.19.0

Compare Source

Ollama is now powered by MLX on Apple Silicon in preview

Ollama on Apple silicon is now built on top of Apple’s machine learning framework, MLX, to take advantage of its unified memory architecture.

mlx-coding-agent.mov

Read more: https://ollama.com/blog/mlx

What's Changed

• Ollama's app will now no longer incorrectly show "model is out of date"
• ollama launch pi now includes web search plugin that uses Ollama's web search
• Improved KV cache hit rate when using the Anthropic-compatible API
• Fixed tool call parsing issue with Qwen3.5 where tool calls would be output in thinking
• MLX runner will now create periodic snapshots during prompt processing
• Fixed KV cache snapshot memory leak in MLX runner
• Fixed issue where flash attention would be incorrectly enabled for grok models
• Fixed qwen3-next:80b not loading in Ollama

New Contributors

• @​amatas made their first contribution in #​15022

Full Changelog: ollama/ollama@v0.18.3...v0.19.0

v0.18.3

Compare Source

Visual Studio Code

Microsoft Visual Studio Code now directly integrates with Ollama via GitHub Copilot.

If you have Ollama installed, any local or cloud model from Ollama can be selected for use within visual studio code.

What's Changed

• GLM parser improvements for tool calls
• OpenClaw integration improvements for gateway checks

Full Changelog: ollama/ollama@v0.18.2...v0.18.3

v0.18.2

Compare Source

What's Changed

• Add extra check to ensure npm and git are installed before installing OpenClaw
• Claude Code will now be faster when run locally, due to preventing cache breakages
• Fix to correctly support ollama launch openclaw --model <model>
• Register Ollama's websearch package correctly for OpenClaw

Full Changelog: ollama/ollama@v0.18.1...v0.18.2

v0.18.1

Compare Source

Web Search and Fetch in OpenClaw

Ollama now ships with web search and web fetch plugin for OpenClaw. This allows Ollama's models (local or cloud) to search the web for the latest content and news. This also allows OpenClaw with Ollama to be able to fetch the web and extract readable content for processing. This feature does not execute JavaScript.

When using local models with web search in OpenClaw, ensure you are signed into Ollama with ollama signin

ollama launch openclaw

You can install web search directly into OpenClaw as a plugin if you already have OpenClaw configured and working:

Ollama web search plugin

openclaw plugins install @​ollama/openclaw-web-search

Non-interactive (headless) mode for ollama launch

ollama launch can now run in non-interactive mode.

Perfect for:

• Docker/containers: spin up an integration as a pipeline step to run evals, test prompts, or validate model behavior as part of your build. Tear it down when the job ends.

• CI/CD: Generate code reviews, security checks, and other tasks within your CI

• Scripts/automation: Kick off automated tasks with Ollama and claude code

• --model must be specified to run in headless mode

• --yes flag will auto-pull the model and skip any selectors

Try with: ollama launch claude --model kimi-k2.5:cloud --yes -- -p "how does this repository work?"

Use non-interactive mode in OpenClaw

You can ask your OpenClaw to run tasks using claude with subagents:

ollama launch claude --model kimi-k2.5:cloud --yes -- -p "how does this repository work?" using a subagent

What's Changed

• ollama launch openclaw will now use the official Ollama auth and model provider for OpenClaw
• Improvements to Ollama's benchmarking tool in ./cmd/bench
• ollama launch openclaw will now skip --install-daemon when systemd is unavailable

Full Changelog: ollama/ollama@v0.18.0...v0.18.1

v0.18.0

Compare Source

Ollama 0.18 includes improved performance for OpenClaw and Ollama’s cloud models, including the new Nemotron-3-Super model by NVIDIA designed for high-performance agentic reasoning tasks.

Improved OpenClaw performance with Kimi-K2.5

This release of Ollama improves performance of cloud models and their reliability.

• Up to 2x faster speeds with Kimi-K2.5
• Tool calling accuracy has been improved

ollama launch openclaw --model kimi-k2.5

Ollama is now a provider in OpenClaw

Ollama can now be selected as an authentication and model provider during OpenClaw onboarding (thanks @​BruceMacD for contributing and @​steipete for reviewing!)

openclaw onboard --auth-choice ollama

More information: https://docs.openclaw.ai/providers/ollama

Nemotron-3-Super

Nemotron-3-Super: is a new 122B parameter model with strong reasoning and tool calling capability, while having top performance when run on modern hardware:

• ollama run nemotron-3-super:cloud
• ollama run nemotron-3-super to run locally (requires 96GB+ of VRAM)

Nemotron-3-Super scores highest of any open model on PinchBench, a benchmark suite that measures how successful models are at completing tasks when used with OpenClaw.

ollama launch openclaw --model nemotron-3-super:cloud

Or using OpenClaw’s onboarding:

openclaw onboard \
	--auth-choice ollama \
	--custom-model-id nemotron-3-super:cloud

Non-interactive task support

ollama launch now supports non-interactive tasks by passing in --yes. This enables using Claude, Codex, Pi and more in scripts, GitHub Actions, and other non-interactive environments.

ollama launch claude \
	--model glm-5:cloud \
	--yes \
	-- "Do a quick code review of this pull request and respond on GitHub with a comment summarizing your feedback."

Lower latency on MiniMax-M2.5 and Qwen3.5 on Ollama’s cloud

For customers in North America, MiniMax-M2.5 and Qwen3.5 on Ollama’s cloud now respond much faster, up to 10x and up to 2x faster respectively, and often in less than a second. This is ideal for tasks that require a fast Time To First Token (TTFT) when needing quick answers from OpenClaw or quick back-to-back coding tasks.

ollama launch claude --model minimax-m2.5

Driver updates required for ROCm 7

This version of Ollama ships with ROCm 7, and requires updating drivers to the latest version for continued support.

What's Changed

• Ollama's cloud models no longer require downloading via ollama pull. Setting :cloud as a tag will now automatically connect to cloud models.
• New --yes flag for ollama launch that skips all prompts, making it possible to run AI assistants and other tools in non-interactive environments
• Fixed issue where "Reset to Defaults" in Ollama's app would disable downloading automatic updates.
• Ollama will now ensure context compaction occurs at the correct context length for each model when using ollama launch claude

New Contributors

• @​flipbit03 made their first contribution in #​14821
• @​shivamtiwari3 made their first contribution in #​14825

Full Changelog: ollama/ollama@v0.17.7...v0.18.0

v0.17.7

Compare Source

What's Changed

• Allow thinking levels such as "medium" to correctly interpreted in Ollama's API for all thinking models
• Add context length to support compaction when using ollama launch

Full Changelog: ollama/ollama@v0.17.6...v0.17.7

v0.17.6

Compare Source

What's Changed

• Fixed issue where GLM-OCR would not work due to incorrect prompt rendering
• Fixed tool calling parsing and rendering for Qwen 3.5 models

New Contributors

• @​Victor-Quqi made their first contribution in #​14584

Full Changelog: ollama/ollama@v0.17.5...v0.17.6

v0.17.5

Compare Source

New models

• Qwen3.5: the small Qwen 3.5 model series is now available in 0.8B, 2B, 4B and 9B parameter sizes.

What's Changed

• Fixed crash in Qwen 3.5 models when split over GPU & CPU
• Fixed issue where Qwen 3.5 models would repeat themselves due to no presence penalty (note: you may have to redownload the qwen3.5 models: ollama pull qwen3.5:35b for example)
• ollama run --verbose will now show peak memory usage when using Ollama's MLX engine
• Fixed memory issues and crashes in MLX runner
• Fixed issue where Ollama would not be able to run models imported from Qwen3.5 GGUF files

Full Changelog: ollama/ollama@v0.17.4...v0.17.5

v0.17.4

Compare Source

New models

• Qwen 3.5: a family of open-source multimodal models that delivers exceptional utility and performance.
• LFM 2: LFM2 is a family of hybrid models designed for on-device deployment. LFM2-24B-A2B is the largest model in the family, scaling the architecture to 24 billion parameters while keeping inference efficient.

Note: for users on 0.17.1, this version will not automatically update. Re-downloading is required to receive the latest version of Ollama.

What's Changed

• Tool call indices will now be included in parallel tool calls

Full Changelog: ollama/ollama@v0.17.3...v0.17.4

v0.17.3

Compare Source

What's Changed

• Fixed issue where tool calls in the Qwen 3 and Qwen 3.5 model families would not be parsed correctly if emitted during thinking

Full Changelog: ollama/ollama@v0.17.2...v0.17.3

v0.17.2

Compare Source

What's Changed

• Fixed issue where Ollama's app on Windows would crash when a new update has been downloaded

Full Changelog: ollama/ollama@v0.17.1...v0.17.2

v0.17.1

Compare Source

What's Changed

• Nemotron architecture support in Ollama's engine
• MLX engine now has improved memory usage
• Ollama's app will now allow models that support tools to use web search capabilities
• Improved LFM2 and LFM2.5 models in Ollama's engine
• ollama create will no longer default to affine quantization for unquantized models when using the MLX engine
• Added configuration for disabling automatic update downloading

Full Changelog: ollama/ollama@v0.17.0...v0.17.1

v0.17.0

Compare Source

OpenClaw

OpenClaw can now be installed and configured automatically via Ollama, making it the easiest way to get up and running with OpenClaw with open models like Kimi-K2.5, GLM-5, and Minimax-M2.5.

Get started

ollama launch openclaw

Web search in OpenClaw

When using cloud models, websearch is enabled - allowing OpenClaw to search the internet.

What's Changed

• Improved tokenizer performance
• Ollama's macOS and Windows apps will now default to a context length based on available VRAM

New Contributors

• @​natl-set made their first contributihttps://github.com/ollama/ollama/pull/14322/14322

Full Changelog: ollama/ollama@v0.16.3...v0.17.0

v0.16.3

Compare Source

What's Changed

• New ollama launch cline added for the Cline CLI
• ollama launch <integration> will now always show the model picker
• Added Gemma 3, Llama and Qwen 3 architectures to MLX runner

New Contributors

• @​hellosaumil made their first contribution in #​14271

Full Changelog: ollama/ollama@v0.16.2...v0.16.3

v0.16.2

Compare Source

What's Changed

• ollama launch claude now supports searching the web when using :cloud models
• Fixed rendering issue when running ollama in PowerShell
• New setting in Ollama's app makes it easier to disable cloud models for sensitive and private tasks where data cannot leave your computer. For Linux or when running ollama serve manually, set OLLAMA_NO_CLOUD=1.
• Fixed issue where experimental image generation models would not run in 0.16.0 and 0.16.1

Full Changelog: ollama/ollama@v0.16.1...v0.16.2-rc0

v0.16.1

Compare Source

What's Changed

• Installing Ollama via the curl install script on macOS will now only prompt for your password if its required
• Installing Ollama via the iem install script in Windows will now show progress
• Image generation models will now respect the OLLAMA_LOAD_TIMEOUT variable

Full Changelog: ollama/ollama@v0.16.0...v0.16.1

v0.16.0

Compare Source

New models

• GLM-5: A strong reasoning and agentic model from Z.ai with 744B total parameters (40B active), built for complex systems engineering and long-horizon tasks.
• MiniMax-M2.5: a new state-of-the-art large language model designed for real-world productivity and coding tasks.

New ollama

The new ollama command makes it easy to launch your favorite apps with models using Ollama

What's Changed

• Launch Pi with ollama launch pi
• Improvements to Ollama's MLX runner to support GLM-4.7-Flash
• Ctrl+G will now allow for editing text prompts in a text editor when running a model

Full Changelog: ollama/ollama@v0.15.6...v0.16.0

v0.15.6

Compare Source

What's Changed

• Fixed context limits when running ollama launch droid
• ollama launch will now download missing models instead of erroring
• Fixed bug where ollama launch claude would cause context compaction when providing images

Full Changelog: ollama/ollama@v0.15.5...v0.15.6

v0.15.5

Compare Source

New models

• Qwen3-Coder-Next: a coding-focused language model from Alibaba's Qwen team, optimized for agentic coding workflows and local development.
• GLM-OCR: GLM-OCR is a multimodal OCR model for complex document understanding, built on the GLM-V encoder–decoder architecture.

Improvements to ollama launch

• ollama launch can now be provided arguments, for example ollama launch claude -- --resume
• ollama launch will now work run subagents when using ollama launch claude
• Ollama will now set context limits for a set of models when using ollama launch opencode

What's Changed

• Sub-agent support for ollama launch for planning, deep research, and similar tasks
• ollama signin will now open a browser window to make signing in easier
• Ollama will now default to the following context lengths based on VRAM:
  • < 24 GiB VRAM: 4,096 context
  • 24-48 GiB VRAM: 32,768 context
  • >= 48 GiB VRAM: 262,144 context
• GLM-4.7-Flash support on Ollama's experimental MLX engine
• ollama signin will now open the browser to the connect page
• Fixed off by one error when using num_predict in the API
• Fixed issue where tokens from a previous sequence would be returned when hitting num_predict

New Contributors

• @​avukmirovich made their first contribution in #​13934

Full Changelog: ollama/ollama@v0.15.4...v0.15.5

v0.15.4

Compare Source

What's Changed

• ollama launch openclaw will now enter the standard OpenClaw onboarding flow if this has not yet been completed.

Full Changelog: ollama/ollama@v0.15.3...v0.15.4

v0.15.3

Compare Source

What's Changed

• Renamed ollama launch clawdbot to ollama launch openclaw to reflect the project's new name
• Improved tool calling for Ministral models
• docs: add clawdbot by @​ParthSareen in #​13925
• cmd/config: Use envconfig.Host() for base API in launch config packages by @​gabe-l-hart in #​13937
• ollama launch will now use the value of OLLAMA_HOST when running it

New Contributors

• @​MBerguer made their first contribution in #​13971
• @​taronsung made their first contribution in #​13965
• @​noureldin-azzab made their first contribution in #​13961
• @​dhirajlochib made their first contribution in #​13645
• @​ThanhNguyxn made their first contribution in #​13979

Full Changelog: ollama/ollama@v0.15.2...v0.15.3

v0.15.2

Compare Source

What's Changed

• New ollama launch clawdbot command for launching Clawdbot using Ollama models

Full Changelog: ollama/ollama@v0.15.1...v0.15.2

v0.15.1

Compare Source

What's Changed

• GLM-4.7-Flash performance and correctness improvements, fixing repetitive answers and tool calling quality
• Fixed performance issues on macOS and arm64 Linux
• Fixed issue where ollama launch would not detect claude and would incorrectly update opencode configurations

New Contributors

• @​stillhart made their first contribution in #​13855

Full Changelog: ollama/ollama@v0.15.0...v0.15.1

v0.15.0

Compare Source

ollama launch

A new ollama launch command to use Ollama's models with Claude Code, Codex, OpenCode, and Droid without separate configuration.

What's Changed

• New ollama launch command for Claude Code, Codex, OpenCode, and Droid
• Fixed issue where creating multi-line strings with """ would not work when using ollama run
• Ctrl+J and Shift+Enter now work for inserting newlines in ollama run
• Reduced memory usage for GLM-4.7-Flash models

v0.14.3

Compare Source

• Z-Image Turbo: 6 billion parameter text-to-image model from Alibaba’s Tongyi Lab. It generates high-quality photorealistic images.
• Flux.2 Klein: Black Forest Labs’ fastest image-generation models to date.

New models

• GLM-4.7-Flash: As the strongest model in the 30B class, GLM-4.7-Flash offers a new option for lightweight deployment that balances performance and efficiency.
• LFM2.5-1.2B-Thinking: LFM2.5 is a new family of hybrid models designed for on-device deployment.

What's Changed

• Fixed issue where Ollama's macOS app would interrupt system shutdown
• Fixed ollama create and ollama show commands for experimental models
• The /api/generate API can now be used for image generation
• Fixed minor issues in Nemotron-3-Nano tool parsing
• Fixed issue where removing an image generation model would cause it to first load
• Fixed issue where ollama rm would only stop the first model in the list if it were running

Full Changelog: ollama/ollama@v0.14.2...v0.14.3

v0.14.2

Compare Source

New models

• TranslateGemma: A new collection of open translation models built on Gemma 3, helping people communicate across 55 languages.

What's Changed

• Shift + Enter (or Ctrl + j) will now enter a newline in Ollama's CLI
• Improve /v1/responses API to better confirm to OpenResponses specification

New Contributors

• @​yuhongsun96 made their first contribution in #​13135
• @​koaning made their first contribution in #​13326

Full Changelog: ollama/ollama@v0.14.1...v0.14.2

v0.14.1

Compare Source

Image generation models (experimental)

Experimental image generation models are available for macOS and Linux (CUDA) in Ollama:

Available models

• Z-Image-Turbo

ollama run x/z-image-turbo

Note: x is a username on ollama.com where experimental models are uploaded

More models coming soon:

1. Qwen-Image-2512
2. Qwen-Image-Edit-2511
3. GLM-Image

What's Changed

• fix macOS auto-update signature verification failure

New Contributors

• @​joshxfi made their first contribution in #​13711
• @​maternion made their first contribution in #​13709

Full Changelog: ollama/ollama@v0.14.0...v0.14.1

v0.14.0

Compare Source

What's Changed

• ollama run --experimental CLI will now open a new Ollama CLI that includes an agent loop and the bash tool
• Anthropic API compatibility: support for the /v1/messages API
• A new REQUIRES command for the Modelfile allows declaring which version of Ollama is required for the model
• For older models, Ollama will avoid an integer underflow on low VRAM systems during memory estimation
• More accurate VRAM measurements for AMD iGPUs
• Ollama's app will now highlight swift source code
• An error will now return when embeddings return NaN or -Inf
• Ollama's Linux install bundles files now use zst compression
• New experimental support for image generation models, powered by MLX

New Contributors

• @​Vallabh-1504 made their first contribution in #​13550
• @​majiayu000 made their first contribution in #​13596
• @​harrykiselev made their first contribution in #​13615

Full Changelog: ollama/ollama@v0.13.5...v0.14.0-rc2

v0.13.5

Compare Source

New Models

• Google's FunctionGemma a specialized version of Google's Gemma 3 270M model fine-tuned explicitly for function calling.

What's Changed

• bert architecture models now run on Ollama's engine
• Added built-in renderer & tool parsing capabilities for DeepSeek-V3.1
• Fixed issue where nested properties in tools may not have been rendered properly

New Contributors

• @​familom made their first contribution in #​13220
• @​nathannewyen made their first contribution in #​13469

Full Changelog: ollama/ollama@v0.13.4...v0.13.5

v0.13.4

Compare Source

New Models

• Nemotron 3 Nano: A new Standard for Efficient, Open, and Intelligent Agentic Models
• Olmo 3 and Olmo 3.1: A series of Open language models designed to enable the science of language models. These models are pre-trained on the Dolma 3 dataset and post-trained on the Dolci datasets.

What's Changed

• Enable Flash Attention automatically for models by default
• Fixed handling of long contexts with Gemma 3 models
• Fixed issue that would occur with Gemma 3 QAT models or other models imported with the Gemma 3 architecture

New Contributors

• @​familom made their first contribution in #​13220

Full Changelog: ollama/ollama@v0.13.3...v0.13.4-rc0

v0.13.3

Compare Source

New models

• Devstral-Small-2: 24B model that excels at using tools to explore codebases, editing multiple files and power software engineering agents.
• rnj-1: Rnj-1 is a family of 8B parameter open-weight, dense models trained from scratch by Essential AI, optimized for code and STEM with capabilities on par with SOTA open-weight models.
• nomic-embed-text-v2: nomic-embed-text-v2-moe is a multilingual MoE text embedding model that excels at multilingual retrieval.

What's Changed

• Improved truncation logic when using /api/embed and /v1/embeddings
• Extend Gemma 3 architecture to support rnj-1 model
• Fix error that would occur when running qwen2.5vl with image input

Full Changelog: ollama/ollama@v0.13.2...v0.13.3

v0.13.2

Compare Source

New models

• Qwen3-Next: The first installment in the Qwen3-Next series with strong performance in terms of both parameter efficiency and inference speed.

What's Changed

• Flash attention is now enabled by default for vision models such as mistral-3, gemma3, qwen3-vl and more. This improves memory utilization and performance when providing images as input.
• Fixed GPU detection on multi-GPU CUDA machines
• Fixed issue where deepseek-v3.1 would always think even with thinking is disabled in Ollama's app

New Contributors

• @​chengcheng84 made their first contribution in #​13265
• @​nathan-hook made their first contribution in #​13256

Full Changelog: ollama/ollama@v0.13.1...v0.13.2

v0.13.1

Compare Source

New models

• Ministral-3: The Ministral 3 family is designed for edge deployment, capable of running on a wide range of hardware.
• Mistral-Large-3: A general-purpose multimodal mixture-of-experts model for production-grade tasks and enterprise workloads.

What's Changed

• nomic-embed-text will now use Ollama's engine by default
• Tool calling support for cogito-v2.1
• Fixed issues with CUDA VRAM discovery
• Fixed link to docs in Ollama's app
• Fixed issue where models would be evicted on CPU-only systems
• Ollama will now better render errors instead of showing Unmarshal: errors
• Fixed issue where CUDA GPUs would fail to be detected with older GPUs
• Added thinking and tool parsing for cogito-v2.1

New Contributors

• @​EntropyYue made their first contribution in #​13237
• @​kokes made their first contribution in #​13231

Full Changelog: ollama/ollama@v0.13.0...v0.13.1

v0.13.0

Compare Source

New models

• DeepSeek-OCR: DeepSeek-OCR uses optical 2D mapping to compress long contexts, achieving high OCR precision with reduced vision tokens and demonstrating practical value in document processing.
• Cogito-V2.1: instruction tuned generative models, currently the best open-weight LLM by a US company

DeepSeek-OCR

DeepSeek-OCR is now available on Ollama. Example inputs:

ollama run deepseek-ocr "/path/to/image\n<|grounding|>Given the layout of the image."

ollama run deepseek-ocr "/path/to/image\nFree OCR."

ollama run deepseek-ocr "/path/to/image\nParse the figure."

ollama run deepseek-ocr "/path/to/image\nExtract the text in the image."

ollama run deepseek-ocr "/path/to/image\n<|grounding|>Convert the document to markdown."

New bench tool

Ollama's GitHub repo now includes a bench tool that can be used to test model performance. For the time being this is a separate tool that can be built in the Ollama GitHub repository:

First, install Go. Then from the root of the Ollama repository run:

go run ./cmd/bench -model gpt-oss:20b

For more information see the tool's documentation

What's Changed

• DeepSeek-OCR is now supported
• DeepSeek-V3.1 architecture is now supported in Ollama's engine
• Fixed performance issues that arose in Ollama 0.12.11 on CUDA
• Fixed issue where Linux install packages were missing required Vulkan libraries
• Improved CPU and memory detection while in containers/cgroups
• Improved VRAM information detection for AMD GPUs
• Improved KV cache performance to no longer require defragmentation

New Contributors

• @​lnicola made their first contribution in #​13096
• @​vignesh1507 made their first contribution in #​13078
• @​pierwill made their first contribution in #​12995
• @​jjuliano made their first contribution in #​11877
• @​omahs made their first contribution in #​10683
• @​SiLeader made their first contribution in #​10292
• @​ssam18 made their first contribution in #​13124
• @​seolyam made their first contribution in #​13116

Full Changelog: ollama/ollama@v0.12.11...v0.13.0

v0.12.11

Compare Source

Logprobs

Ollama's API and OpenAI-compatible API now support log probabilities. Log probabilities of output tokens indicate the likelihood of each token occurring in the sequence given the context. This is useful for different use cases:

1. Classification tasks
2. Retrieval (Q&A) evaluation
3. Autocomplete
4. Token highlighting and outputting bytes
5. Calculating perplexity

To enable Logprobs, provide "logprobs": true to Ollama's API:

curl http://localhost:11434/api/generate -d '{
  "model": "gemma3",
  "prompt": "Why is the sky blue?",
  "logprobs": true
}'

When log probabilities are requested, response chunks will now include a "logprobs" field with the token, log probability and raw bytes (for partial unicode).

{
  "model": "gemma3",
  "created_at": "2025-11-14T22:17:56.598562Z",
  "response": "Okay",
  "done": false,
  "logprobs": [
    {
      "token": "Okay",
      "logprob": -1.3434503078460693,
      "bytes": [
        79,
        107,
        97,
        121
      ]
    }
  ]
}

top_logprobs

When setting "top_logprobs", a number of most-likely tokens are also provided, making it possible to introspect alternative tokens. Below is an example request.

curl http://localhost:11434/api/generate -d '{
  "model": "gemma3",
  "prompt": "Why is the sky blue?",
  "logprobs": true,
  "top_logprobs": 3
}'

This will generate a stream of response chunks with the following fields:

{
  "model": "gemma3",
  "created_at": "2025-11-14T22:26:10.466324Z",
  "response": "The",
  "done": false,
  "logprobs": [
    {
      "token": "The",
      "logprob": -0.8361086845397949,
      "bytes": [
        84,
        104,
        101
      ],
      "top_logprobs": [
        {
          "token": "The",
          "logprob": -0.8361086845397949,
          "bytes": [
            84,
            104,
            101
          ]
        },
        {
          "token": "Okay",
          "logprob": -1.2590975761413574,
          "bytes": [
            79,
            107,
            97,
            121
          ]
        },
        {
          "token": "That",
          "logprob": -1.2686877250671387,
          "bytes": [
            84,
            104,
            97,
            116
          ]
        }
      ]
    }
  ]
}

Special thanks

Thank you @​baptistejamin for adding Logprobs to Ollama's API.

Vulkan support (opt-in)

Ollama 0.12.11 includes support for Vulkan acceleration. Vulkan brings support for a broad range of GPUs from AMD, Intel, and iGPUs. Vulkan support is not yet enabled by default, and requires opting in by running Ollama with a custom environment variable:

OLLAMA_VULKAN=1 ollama serve

On Powershell, use:

$env:OLLAMA_VULKAN="1"
ollama serve

For issues or feedback on using Vulkan with Ollama, create an issue labelled Vulkan and make sure to include server logs where possible to aid in debugging.

What's Changed

• Ollama's API and the OpenAI-compatible API now supports Logprobs
• Ollama's new app now supports WebP images
• Improved rendering performance in Ollama's new app, especially when rendering code
• The "required" field in tool definitions will now be omitted if not specified
• Fixed issue where "tool_call_id" would be omitted when using the OpenAI-compatible API.
• Fixed issue where ollama create would import data from both consolidated.safetensors and other safetensor files.
• Ollama will now prefer dedicated GPUs over iGPUs when scheduling models
• Vulkan can now be enabled by setting OLLAMA_VULKAN=1. For example: OLLAMA_VULKAN=1 ollama serve

New Contributors

• @​mags0ft made their first contribution in #​11371
• @​macarronesc made their first contribution in #​12973
• @​breatn made their first contribution in #​12985
• @​cybardev made their first contribution in #​13045
• @​baptistejamin made their first contribution in #​12899

Full Changelog: ollama/ollama@v0.12.10...v0.12.11

v0.12.10

Compare Source

ollama run now works with embedding models

ollama run can now run embedding models to generate vector embeddings from text:

ollama run embeddinggemma "Hello world"

Content can also be provided to ollama run via standard input:

echo "Hello world" | ollama run embeddinggemma

What's Changed

• Fixed errors when running qwen3-vl:235b and qwen3-vl:235b-instruct
• Enable flash attention for Vulkan (currently needs to be built from source)
• Add Vulkan memory detection for Intel GPU using DXGI+PDH
• Ollama will now return tool call IDs from the /api/chat API
• Fixed hanging due to CPU discovery
• Ollama will now show login instructions when switching to a cloud model in interactive mode
• Fix reading stale VRAM data
• ollama run now works with embedding models

New Contributors

• @​ryanycoleman made their first contribution in #​11740
• @​Rajathbail made their first contribution in #​12929
• @​virajwad made their first contribution in #​12664
• @​AXYZdong made their first contribution in #​8601

Full Changelog: ollama/ollama@v0.12.9...v0.12.10

v0.12.9

Compare Source

What's Changed

• Fix performance regression on CPU-only systems

Full Changelog: ollama/ollama@v0.12.8...v0.12.9

v0.12.8

Compare Source

What's Changed

• qwen3-vl performance improvements, including flash attention support by default
• qwen3-vl will now output less leading whitespace in the response when thinking
• Fixed issue where deepseek-v3.1 thinking could not be disabled in Ollama's new app
• Fixed issue where qwen3-vl would fail to interpret images with transparent backgrounds
• Ollama will now stop running a model before removing it via ollama rm
• Fixed issue where prompt processing would be slower on Ollama's engine
• Ignore unsupported iGPUs when doing device discovery on Windows

New Contributors

• @​athshh made their first contributihttps://github.com/ollama/ollama/pull/12822/12822

Full Changelog: ollama/ollama@v0.12.7...v0.12.8

v0.12.7

Compare Source

<img width="600" alt="Ollama screenshot 2025-10-29

---

Configuration

📅 Schedule: (UTC)

• Branch creation
  • At any time (no schedule defined)
• Automerge
  • At any time (no schedule defined)

🚦 Automerge: Disabled by config. Please merge this manually once you are satisfied.

♻ Rebasing: Whenever PR becomes conflicted, or you tick the rebase/retry checkbox.

🔕 Ignore: Close this PR and you won't be reminded about this update again.

---

• If you want to rebase/retry this PR, check this box

---

This PR was generated by Mend Renovate. View the repository job log.

[renovate]

ℹ Artifact update notice

File name: go.mod

In order to perform the update(s) described in the table above, Renovate ran the go get command, which resulted in the following additional change(s):

• 6 additional dependencies were updated
• The go directive was updated for compatibility reasons

Details:

Package	Change
go	1.23.0 -> 1.24.1
golang.org/x/crypto	v0.26.0 -> v0.33.0
golang.org/x/exp	v0.0.0-20240808152545-0cdaa3abc0fa -> v0.0.0-20250218142911-aa4b98e5adaa
golang.org/x/sync	v0.8.0 -> v0.11.0
golang.org/x/net	v0.27.0 -> v0.35.0
golang.org/x/sys	v0.23.0 -> v0.30.0
golang.org/x/text	v0.17.0 -> v0.22.0

[renovate]

ℹ️ Artifact update notice

File name: go.mod

In order to perform the update(s) described in the table above, Renovate ran the go get command, which resulted in the following additional change(s):

• 6 additional dependencies were updated
• The go directive was updated for compatibility reasons

Details:

Package	Change
go	1.23.0 -> 1.24.1
golang.org/x/crypto	v0.26.0 -> v0.43.0
golang.org/x/exp	v0.0.0-20240808152545-0cdaa3abc0fa -> v0.0.0-20250218142911-aa4b98e5adaa
golang.org/x/sync	v0.8.0 -> v0.17.0
golang.org/x/net	v0.27.0 -> v0.46.0
golang.org/x/sys	v0.23.0 -> v0.37.0
golang.org/x/text	v0.17.0 -> v0.30.0