[mchesterbot]

This PR contains the following updates:

Package	Type	Update	Change
@ai-sdk/anthropic (source)	dependencies	patch	4.0.90 → 4.0.94

---

Release Notes

vercel/ai (@​ai-sdk/anthropic)

v4.0.94

Compare Source

Patch Changes

• f4c5f1e: fix(anthropic): use native structured outputs for Claude models that reject forced tool use (Sonnet 5.5, Opus 5.5, Fable 5.1) even when the provider disables them by default, as Amazon Bedrock and Google Vertex do. Self-referencing schemas keep the JSON response tool. Also keep definitions and $defs next to a top-level $ref when sanitizing structured output schemas, so the reference still resolves.
• e10aebb: fix(anthropic): when the JSON response tool is in use but the model answers in text without calling it (for example with manual thinking, where the forced tool choice is downgraded to auto), return the final answer text instead of dropping it. Text before a tool call, result, or message continuation is still dropped.
• b0d51fe: fix(anthropic): keep the JSON response tool for schemas that native structured outputs cannot represent when upgrading models that reject forced tool use. Schemas with open objects (maps, free-form objects) would otherwise come back as {}, and schemas with more than 10 optional properties can be rejected as too complex.

v4.0.93

Compare Source

Patch Changes

• dcd2f8d: Introduce a model entry contract with named capability sections and request configuration. Consolidate the OpenAI and Anthropic model catalogs so existing request behavior and generated capability documentation use the same source of truth. Catalogs also declare the provider options each provider accepts and which models support them.
• Updated dependencies [7716e18]
• Updated dependencies [5c247e9]
• Updated dependencies [dcd2f8d]
  • @​ai-sdk/provider-utils@5.0.72
  • @​ai-sdk/provider@4.0.35

v4.0.92

Compare Source

Patch Changes

• 61a315e: fix: map the top-level reasoning option consistently across providers

  Providers now resolve reasoning the same way:

  1. A provider-specific reasoning setting in providerOptions is sent unchanged. OpenAI no longer drops providerOptions.openai.reasoningEffort values that are not in its list for the model, and Fireworks no longer rewrites providerOptions.fireworks.reasoningEffort. Fireworks also sends top-level reasoning values directly instead of rounding them to 'high' at most, except for model families known to reject 'none' (GPT-OSS, GLM-5.3, Muse Spark) or 'minimal' (Nemotron, Inkling).
  2. Otherwise, reasoning maps to the closest level the model supports, rounding down, and rounds up only when there is no lower level. For example, 'xhigh' now maps to 'high' instead of 'max' on Anthropic and Bedrock models without xhigh effort, and 'medium' maps to 'low' instead of 'high' on DeepSeek, Kimi K3, and Mistral GLM-5.3 models.
  3. 'none' on a model that cannot disable reasoning maps to the model's lowest level with a compatibility warning, instead of being dropped or sending a request the API rejects. This affects GPT-6 models without 'none', xAI grok-4.5 and later, Groq GPT-OSS, Kimi K3, Perplexity, QuiverAI, Bedrock OpenAI gpt-oss models, and Gemini 2.5 Pro (minimum thinking budget). Gemini 3 and Anthropic and Bedrock between_tools mappings now emit the compatibility warning too.

  mapReasoningToProviderEffort now accepts any reasoning value, returns an optional providerOptionsEffort unchanged, and rounds levels that are missing from the effort map to the closest supported level. mapReasoningToProviderBudget maps 'none' to the minimum reasoning budget, always returns a budget, and uses the default percentage for levels missing from budgetPercentages. OpenAICompatibleChatConfig adds an optional mapReasoning function for mapping the top-level reasoning option.

• Updated dependencies [61a315e]

  • @​ai-sdk/provider-utils@5.0.71

v4.0.91

Compare Source

Patch Changes

• 50da4c5: Fix workflow serialization of Anthropic models using API keys or synchronous auth tokens while preserving support for asynchronous token providers.

---

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 has been generated by Mend Renovate CLI.

[claude]

Claude finished @mchesterbot[bot]'s task in 18s —— View job

---

Review of the @ai-sdk/anthropic bump

• Read the diff (package.json, package-lock.json)
• Checked usage in the codebase and the lockfile tree

Summary: This is a patch bump with a small blast radius. The only consumer is createAnthropic in lib/llm/registry.ts:3. I did not run the tests or CI.

Notes

• Title is stale. The PR title and description say 4.0.93, but the branch (785c09d) pins 4.0.94. The release notes cover only 4.0.91 to 4.0.93, so 4.0.94 is not described. Renovate will probably refresh the text, or you can retitle it.
• Duplicate transitive packages. @ai-sdk/anthropic@4.0.94 needs @ai-sdk/provider@4.0.35 and provider-utils@5.0.72. The other @ai-sdk/* packages still resolve to 4.0.34 and 5.0.70 at the top level. npm therefore nests a second copy of both, plus undici@7.30.1, under node_modules/@ai-sdk/anthropic/. It's harmless at runtime, but two copies of @ai-sdk/provider can cause structural type mismatches when an Anthropic model is passed to ai helpers. Bumping ai, @ai-sdk/google, @ai-sdk/openai and @ai-sdk/openai-compatible together would dedupe the tree. If npm run typecheck is green, this isn't blocking.
• Behavior change to watch. The 4.0.92 changelog changes how the top-level reasoning option maps on Anthropic. For example, 'xhigh' now rounds down to 'high' on models without xhigh effort, where it used to go to 'max'. 4.0.93 also changes the model catalog. If lib/llm/ passes reasoning or effort values, or the model table in lib/llm/models.ts assumes particular catalog behavior, it's worth a quick check. I only grepped for the import and did not trace the options.
• Changelog fragment. None needed. The change touches no migrations, env, Dockerfile or deploy files.
• Security. Nothing concerning. The new undici is a patch-level 7.x release.

Verdict: Looks fine to merge once Typecheck, Unit Tests and Build pass. Check the reasoning-mapping change above if you set reasoning effort for Anthropic models.
· branch renovate/ai-sdk-anthropic-4.x