@ai-sdk/openai-compatible
3.0.93 · 17 Nov 2026Apache-2.014M / weekRepositoryHomepagenpmAtom feed
1.0.63 is the newest release of 1.x.
In short
2 breaking7 new8 fixed5 internal
- Breaking
AI SDK 5
- Breaking
Release note
Major Changes
- Breaking
AI SDK 5
- Breaking
Move Image Model Settings into generate options
Image Models no longer have settings. Instead,
maxImagesPerCallcan be passed directly togenerateImage(). All other image settings can be passed toproviderOptions[provider].Before
await generateImage({ model: luma.image('photon-flash-1', { maxImagesPerCall: 5, pollIntervalMillis: 500, }), prompt, n: 10, });After
await generateImage({ model: luma.image('photon-flash-1'), prompt, n: 10, maxImagesPerCall: 5, providerOptions: { luma: { pollIntervalMillis: 5 }, }, });Pull Request: #6180
Patch Changes
providers/openai-compatible: Convert to providerOptions
providers/xai: Return actual usage when streaming instead of NaN
Fix (provider/openai-compatible): change tool_call type schema to nullish
Add raw chunk support
embedding-model-v2: Add response body field
providers/xai: Add reasoningEffort provider option
Allow any string as reasoningEffort
packages: Export node10 compatible types
Fix (image-model):
specificationVersion: v1->v2Refactor (image-model): rename
ImageModelV1toImageModelV2Add (optional) includeUsage option to createOpenAICompatible
Using Zod 4 for internal stuff
Feat (provider/openai-compatible): allow providers that do not specify function type
Improve Zod compatibility
Fix (provider/openai-incompatible): support empty tool calls
A whole project to check? Paste its package.json in the dependency check: every deprecated, outdated or vulnerable dependency, with what to upgrade to. Free, no account.