[mchesterbot]

This PR contains the following updates:

Package	Type	Update	Change
@ai-sdk/openai-compatible (source)	dependencies	patch	3.0.80 → 3.0.84

---

Release Notes

vercel/ai (@​ai-sdk/openai-compatible)

v3.0.84

Compare Source

Patch Changes

• Updated dependencies [fa15312]
  • @​ai-sdk/provider-utils@5.0.74

v3.0.83

Compare Source

Patch Changes

• 4ae6986: fix(openai-compatible): preserve string tool call arguments and safely serialize BigInt and circular structures in tool calls and outputs
• 5006d83: fix(openai-compatible): send reasoningEffort: null as "reasoning_effort": null, and stop an unset reasoningEffort from overwriting a raw reasoning_effort provider option
• Updated dependencies [51ca5b7]
• Updated dependencies [dc6f1c3]
• Updated dependencies [4ae6986]
  • @​ai-sdk/provider-utils@5.0.73

v3.0.82

Compare Source

Patch Changes

• 7716e18: Support promises and synchronous or asynchronous callbacks in provider API key and token settings. Resolve credentials per authenticated request while preserving existing authentication behavior and synchronous credential serialization.

  The experimental OpenAI Live getServerWebSocketConfig() method can now resolve asynchronously. Await its result before using the connection settings.

• Updated dependencies [7716e18]

• Updated dependencies [5c247e9]

• Updated dependencies [dcd2f8d]

  • @​ai-sdk/provider-utils@5.0.72
  • @​ai-sdk/provider@4.0.35

v3.0.81

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

---

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 15s —— View job

---

Review: @ai-sdk/openai-compatible 3.0.80 → 3.0.84

Verdict: looks safe to merge once CI is green.

• The PR is a pin bump in package.json plus the matching package-lock.json change. The branch has since moved on to 3.0.84 (the title and description still say 3.0.83). The lockfile and package.json agree.
• Transitive bumps: @ai-sdk/provider 4.0.35, @ai-sdk/provider-utils 5.0.74, undici ^7.29.1. All are nested under the package's own node_modules, so the top-level @ai-sdk/* packages are untouched.
• Integrity hashes are in place. There are no unexpected new dependencies.
• Behaviour changes in the release notes (3.0.81–3.0.83):
  • Reasoning mapping is more consistent. This only matters for reasoning / reasoningEffort.
  • reasoningEffort: null is now sent explicitly.
  • Tool-call string arguments are preserved, and BigInt or circular structures serialize safely.
  • API keys can now be async.
• Impact on this repo: the openai-compatible provider is used through lib/llm/registry.ts. lib/llm/models.ts reports no structured-output support and no effort mapping for it (resolveEffort returns null). I didn't find the app passing reasoningEffort through this provider, so the reasoning changes should have no effect here. The tool-call serialization fix could help the admin assistant when it runs on a local server such as Ollama.
• I didn't run tests locally. lib/llm/__tests__/registry.test.ts covers building the compatible chat model, so CI's Unit Tests, Typecheck and Build should be enough to confirm the bump.
• No changelog fragment is needed. This doesn't touch prisma/migrations, lib/env.ts or the Docker and deploy files.

No code issues found.