chore(cognitive): update AI model catalog

[github-actions]

Model Update Summary

Updated 22 models across 6 providers (5 new Anthropic, 3 new Google AI, 4 new/updated Groq, 2 new Cerebras, 2 new xAI, 2 new Fireworks AI; 3 existing model updates).

---

Anthropic

New Models Added

Model ID	Change	Notes
claude-opus-4-6	NEW	Latest flagship model. $5/$25 per 1M tokens, 200K context, 128K max output, lifecycle: production
claude-sonnet-4-6	NEW	Latest balanced model. $3/$15 per 1M tokens, 200K context, 64K max output, lifecycle: production
claude-opus-4-5-20251101	NEW	Legacy production model (alias: claude-opus-4-5). $5/$25 per 1M tokens, 200K context, 64K max output
claude-opus-4-1-20250805	NEW	Legacy production model (alias: claude-opus-4-1). $15/$75 per 1M tokens, 200K context, 32K max output
claude-opus-4-20250514	NEW	First Claude 4 Opus model (alias: claude-opus-4-0). $15/$75 per 1M tokens, 200K context, 32K max output

Existing Model Updates

Model ID	Field	Old Value	New Value
claude-3-haiku-20240307	lifecycle	production	deprecated
claude-3-haiku-20240307	deprecationDate	(none)	2026-04-19
claude-3-haiku-20240307	discontinuedDate	(none)	2026-04-19
claude-3-haiku-20240307	replacementModels	(none)	['claude-haiku-4-5-20251001']
claude-3-haiku-20240307	tags	['low-cost', 'general-purpose']	['deprecated', 'low-cost', 'general-purpose']

Config Changes

• defaultModel: claude-sonnet-4-5-20250929 → claude-sonnet-4-6

Source: https://docs.anthropic.com/en/docs/about-claude/models

---

OpenAI

No changes — official documentation returned HTTP 403. Existing models in config appear current based on available data.

---

Google AI

New Models Added

Model ID	Change	Notes
gemini-3.1-pro	NEW	internalModelId: gemini-3.1-pro-preview. $2/$12 per 1M tokens (≤200K), 1M context, 65K max output, lifecycle: preview
gemini-3.1-flash-lite	NEW	internalModelId: gemini-3.1-flash-lite-preview. $0.25/$1.50 per 1M tokens, 1M context, 65K max output, lifecycle: preview
gemini-2.5-flash-lite	NEW	$0.10/$0.40 per 1M tokens, 1M context, 65K max output, lifecycle: production

Existing Model Updates

No pricing or capability changes confirmed for existing models (gemini-2.5-flash, gemini-2.5-pro, gemini-2.0-flash, gemini-3-pro, gemini-3-flash remain unchanged).

Source: https://ai.google.dev/gemini-api/docs/models, https://ai.google.dev/pricing

---

Groq

Existing Model Updates

Model ID	Field	Old Value	New Value
gpt-oss-20b	inputCostPer1mTokens	0.1	0.075
gpt-oss-20b	outputCostPer1mTokens	0.5	0.3
gpt-oss-20b	maxOutputTokens	32_000	65_536
gpt-oss-120b	outputCostPer1mTokens	0.75	0.6
gpt-oss-120b	maxOutputTokens	32_000	65_536

New Models Added

Model ID	Change	Notes
kimi-k2-instruct-0905	NEW	internalModelId: moonshotai/kimi-k2-instruct-0905. $1.00/$3.00 per 1M tokens, 262K context, 16K max output, lifecycle: production (preview tag)
qwen3-32b	NEW	internalModelId: qwen/qwen3-32b. $0.29/$0.59 per 1M tokens, 131K context, 40K max output, lifecycle: production (preview tag)

Source: https://console.groq.com/docs/models

---

Cerebras

New Models Added

Model ID	Change	Notes
qwen-3-235b-a22b-instruct-2507	NEW	Qwen3 235B MoE (22B active). $0.60/$1.20 per 1M tokens, 32K context, 16K max output, lifecycle: production (preview)
zai-glm-4.7	NEW	Z.ai GLM 4.7 355B. $2.25/$2.75 per 1M tokens, 8K context, 8K max output, lifecycle: production (preview)

Note: Context window sizes for new Cerebras models are estimates — the official docs do not publish these numbers.

Source: https://www.cerebras.ai/pricing, https://inference-docs.cerebras.ai/models/overview

---

xAI

New Models Added

Model ID	Change	Notes
grok-4-1-fast-reasoning	NEW	$0.20/$0.50 per 1M tokens, 2M context, 128K max output, reasoning + vision, lifecycle: production
grok-4-1-fast-non-reasoning	NEW	$0.20/$0.50 per 1M tokens, 2M context, 128K max output, vision, lifecycle: production

Source: https://docs.x.ai/docs/models

---

OpenRouter

No changes — existing gpt-oss-120b model is current. OpenRouter renders models dynamically; no additional top-model additions made beyond confirmed data.

---

Fireworks AI

New Models Added

Model ID	Change	Notes
kimi-k2-instruct	NEW	internalModelId: accounts/fireworks/models/kimi-k2-instruct. $0.60/$2.50 per 1M tokens, 262K context, 16K max output, lifecycle: production
deepseek-v3p1	NEW	internalModelId: accounts/fireworks/models/deepseek-v3p1. $0.56/$1.68 per 1M tokens, 163K context, 16K max output, lifecycle: production

Source: https://fireworks.ai/models, https://fireworks.ai/pricing

[allardy]

/models remove fireworks changes, ceerebras, grow

[github-actions]

No changes were needed for this instruction. The config files already match, or the instruction could not be applied.

Claude output

[allardy]

/models remove fireworks changes, ceerebras, grow

[github-actions]

Applied the adjustment and pushed a new commit.