@ai-sdk/openai-compatible
3.0.93 · 17 Nov 2026Apache-2.014M / weekRepositoryHomepagenpmAtom feed
299 stable releases, newest first, by release line. Notes are quoted as the maintainers published them.
- 3.x · latest line
3.0.41
30 Aug 2026Support text and thinking parts in array-based chat completion content while ignoring unknown part types.
Patch Changes
Support text and thinking parts in array-based chat completion content while ignoring unknown part types.
3.0.37 – 3.0.403 releases that only update dependencies@ai-sdk/provider-utils 5.0.33@ai-sdk/provider 4.0.8Show each
- 3.0.4028 Aug 2026@ai-sdk/provider-utils 5.0.33
- 3.0.3926 Aug 2026@ai-sdk/provider-utils 5.0.32
- 3.0.3725 Aug 2026@ai-sdk/provider 4.0.8@ai-sdk/provider-utils 5.0.30
3.0.36
24 Aug 2026Send
reasoning_effort: "none"when top-level reasoning is disabled
3.0.35
22 Aug 2026openai-compatible: Convert video file parts to
video_urlcontent parts
Patch Changes
openai-compatible: Convert video file parts to
video_urlcontent parts
3.0.34
21 Aug 2026provider/openai-compatible: Report image generation token usage
3.0.33
20 Aug 2026openai-compatible: Report truncated chat streams as errors
3.0.32
19 Aug 2026Handle empty string tool call IDs
Preserve Gemini thought signatures for custom OpenAI-compatible provider names
Patch Changes
Handle empty string tool call IDs
Preserve Gemini thought signatures for custom OpenAI-compatible provider names
3.0.31
17 Aug 2026openai-compatible: Preserve unmapped usage fields in
usage.rawusage.rawis specified as usage "in the shape that the provider returns", and the chat model already parsed the usage object loosely so that extra top-level fields survived. The nestedprompt_tokens_detailsandcompletion_tokens_detailsobjects were still strict, so anything a provider reported inside them was dropped — which is where providers put their most distinguishing detail. The existing "should preserve extra usage fields" fixture was itself losingaudio_tokens,image_tokensandtext_tokensthis way.Both nested objects are now parsed loosely, as is the completion model's usage schema, which was strict throughout despite feeding the same
rawfield.Only
usage.rawchanges. The mapped token counts are unaffected.
3.0.29 – 3.0.302 releases that only update dependencies@ai-sdk/provider-utils 5.0.27Show each
- 3.0.3011 Aug 2026@ai-sdk/provider-utils 5.0.27
- 3.0.2910 Aug 2026@ai-sdk/provider-utils 5.0.26
3.0.28
9 Aug 2026provider/openai-compatible: Clamp
outputTokens.textat 0 when a provider reportscompletion_tokens_details.reasoning_tokensgreater thancompletion_tokens(observed with Baseten serving reasoning models that hit the length stop mid-reasoning). The text share of completion tokens can never be negative;totalandreasoningremain as reported by the provider.
3.0.24 – 3.0.274 releases that only update dependencies@ai-sdk/provider 4.0.7@ai-sdk/provider-utils 5.0.25Show each
3.0.23
4 Aug 2026Fix streamed tool calls with non-zero, non-contiguous, reused, or missing indexes.
Patch Changes
Fix streamed tool calls with non-zero, non-contiguous, reused, or missing indexes.
3.0.14 – 3.0.229 releases that only update dependencies@ai-sdk/provider-utils 5.0.20@ai-sdk/provider 4.0.5Show each
- 3.0.224 Aug 2026@ai-sdk/provider-utils 5.0.20
- 3.0.213 Aug 2026@ai-sdk/provider 4.0.5@ai-sdk/provider-utils 5.0.19
- 3.0.2031 Jul 2026@ai-sdk/provider-utils 5.0.18
- 3.0.1931 Jul 2026@ai-sdk/provider-utils 5.0.17
- 3.0.1830 Jul 2026@ai-sdk/provider-utils 5.0.16
- 3.0.1729 Jul 2026@ai-sdk/provider-utils 5.0.15
- 3.0.1628 Jul 2026@ai-sdk/provider-utils 5.0.14
- 3.0.1527 Jul 2026@ai-sdk/provider 4.0.4@ai-sdk/provider-utils 5.0.13
- 3.0.1421 Jul 2026@ai-sdk/provider-utils 5.0.12
3.0.13
20 Jul 2026openai-compatible: Add an extensible
providerOptionstype for image model requests and stop forcingresponse_format
3.0.12
17 Jul 2026Only updates a dependency:@ai-sdk/provider-utils 5.0.11
3.0.11
15 Jul 2026Preserve structured error data from chat completion SSE streams.
3.0.7 – 3.0.103 releases that only update dependencies@ai-sdk/provider-utils 5.0.10@ai-sdk/provider 4.0.3Show each
- 3.0.1014 Jul 2026@ai-sdk/provider-utils 5.0.10
- 3.0.913 Jul 2026@ai-sdk/provider-utils 5.0.9
- 3.0.79 Jul 2026@ai-sdk/provider 4.0.3@ai-sdk/provider-utils 5.0.7
3.0.6
8 Jul 2026Fix
StreamingToolCallTrackerfinalizing streaming tool calls on parsable partial JSON. Tool calls now only finalize during stream flush, restoring the behavior of #13137: a parsable argument buffer can still be the prefix of a longer argument string, so finalizing early could act on truncated tool inputs.
Patch Changes
Fix
StreamingToolCallTrackerfinalizing streaming tool calls on parsable partial JSON. Tool calls now only finalize during stream flush, restoring the behavior of #13137: a parsable argument buffer can still be the prefix of a longer argument string, so finalizing early could act on truncated tool inputs.
3.0.5
2 Jul 2026Add experimental streaming transcription support for transcription models, including OpenAI
gpt-realtime-whisperand xAI WebSocket STT.
Patch Changes
Add experimental streaming transcription support for transcription models, including OpenAI
gpt-realtime-whisperand xAI WebSocket STT.
3.0.1 – 3.0.44 releases that only update dependencies@ai-sdk/provider-utils 5.0.4@ai-sdk/provider 4.0.1Show each
- 3.0.42 Jul 2026@ai-sdk/provider-utils 5.0.4
- 3.0.31 Jul 2026@ai-sdk/provider-utils 5.0.3
- 3.0.230 Jun 2026@ai-sdk/provider 4.0.1@ai-sdk/provider-utils 5.0.2
- 3.0.127 Jun 2026@ai-sdk/provider-utils 5.0.1
3.0.0
25 Jun 2026- Breaking
Remove CommonJS exports from all packages. All packages are now ESM-only (
"type": "module"). Consumers usingrequire()must switch to ESMimportsyntax. - Breaking
provider: Add support for provider references and uploading files as supported per provider
- Breaking
Start v7 pre-release
Major Changes
- Breaking
Remove CommonJS exports from all packages. All packages are now ESM-only (
"type": "module"). Consumers usingrequire()must switch to ESMimportsyntax. - Breaking
provider: Add support for provider references and uploading files as supported per provider
- Breaking
Start v7 pre-release
- Breaking
Make provider implementations code patterns more consistent, including renaming certain exported symbols
For all externally exported symbols that were renamed, the old names continue to work via deprecated aliases.
Patch Changes
openai-compatible: Buffer tool call deltas until function.name arrives
Add AI Gateway hint to provider READMEs
Migrate providers to support new top-level
reasoningparametervertex: Add grok models to vertex provider
openai-compatible: Honor camelCase providerOptions key in chat and completion models
Accept empty string
rolein streaming delta chunks from OpenAI-compatible providerssecurity: Prevent streaming tool calls from finalizing on parsable partial JSON
Streaming tool call arguments were finalized using
isParsableJson()as a heuristic for completion. If partial accumulated JSON happened to be valid JSON before all chunks arrived, the tool call would be executed with incomplete arguments. Tool call finalization now only occurs inflush()after the stream is fully consumed.Trigger release for all packages after provenance setup
More precise default message for tool execution denial
Revert incorrect fix #13172
openai-compatible: Emit warning when using kebab-case instead of camelCase
- Breaking
Raise minimum supported Node.js version to 22. Supported versions: 22, 24, and 26.
Extract shared
StreamingToolCallTrackerclass into@ai-sdk/provider-utilsto deduplicate streaming tool call handling across OpenAI-compatible providers. Also adds missinggenerateId()fallback fortoolCallIdin Alibaba'sdoGeneratepath and ensures all providers finalize unfinished tool calls during stream flush.Trigger initial canary release
provider: Preserve error type prefix in getErrorMessage
openai, openai-compatible: Only send null content for assistant messages with tool calls
provider: Change file part data property to be tagged with a type and remove the image part type
Trigger initial beta release
Patch - send content: null instead of empty string for tool-only assistant messages
Add workflow serialization support to all provider models.
@ai-sdk/provider-utils: NewserializeModel()helper that extracts only serializable properties from a model instance, filtering out functions and objects containing functions. Third-party provider authors can use this to add workflow support to their own models.All providers:
headersis now optional in provider config types. This is non-breaking — existing code that passesheaderscontinues to work. Custom provider implementations that construct model configs manually can now omitheaders, which is useful when models are deserialized from a workflow step boundary where auth is provided separately.All provider model classes now include
WORKFLOW_SERIALIZEandWORKFLOW_DESERIALIZEstatic methods, enabling them to cross workflow step boundaries without serialization errors.
- 2.x
2.0.92 – 2.0.943 releases that only update dependencies@ai-sdk/provider-utils 4.0.67Show each
- 2.0.9417 Nov 2026@ai-sdk/provider-utils 4.0.67
- 2.0.9310 Nov 2026@ai-sdk/provider-utils 4.0.66
- 2.0.927 Nov 2026@ai-sdk/provider-utils 4.0.65
2.0.91
6 Nov 2026Avoid empty structured-output answers from Baseten's DeepSeek V4 Pro 0813 when reasoning is enabled. Fall back to JSON prompt instructions while keeping reasoning and SDK output validation, and report the fallback through request transformation warnings.
openai,openai-compatible: Surface refusal logprobs and openai-compatible refusals
@ai-sdk/openai: Chat Completions responses that contain a refusal can returnlogprobs.refusalnext tologprobs.content. The provider previously read onlylogprobs.content, soproviderMetadata.openai.logprobswas empty or missing for refusals. Refusal logprobs are now included in emission order (content logprobs followed by refusal logprobs) in bothgenerateTextandstreamText.Streaming logprobs are now collected across all chunks in
streamText: previously only the logprobs of the last streamed chunk were reported. This applies to content logprobs as well, soproviderMetadata.openai.logprobsnow contains the logprobs of every streamed token instead of only the last chunk's tokens. Explicitly empty logprobs (logprobs: { content:, refusal: null }) are reported as an empty array, as before.A refusal that arrives in the same streamed chunk as the
finish_reasonis now taken into account: such a response is reported with acontent-filterfinish reason instead ofstop.@ai-sdk/openai-compatible: Chat Completions responses that carrymessage.refusal(non-streaming) ordelta.refusal(streaming) were ignored, so providers such as OpenAI served throughcreateOpenAICompatiblelost the refusal text and reported an emptystopresponse. Refusals are now returned as text content after any content text, and a non-empty refusal with astopfinish reason is reported as acontent-filterfinish reason. This also applies when the refusal and the finish reason arrive in the same streamed chunk. Non-string refusals (for examplerefusal: falsefrom servers that use booleans) are ignored instead of failing the response.
2.0.90
3 Nov 2026openai-compatible: Preserve string tool call arguments and safely serialize BigInt and circular structures in tool calls and outputs
Patch Changes
openai-compatible: Preserve string tool call arguments and safely serialize BigInt and circular structures in tool calls and outputs
2.0.89
2 Nov 2026openai-compatible: Send
reasoningEffort: nullas"reasoning_effort": null, and stop an unsetreasoningEffortfrom overwriting a rawreasoning_effortprovider option
2.0.88
29 Oct 2026Only updates a dependency:@ai-sdk/provider-utils 4.0.63
2.0.87
21 Oct 2026Defer validation schema construction until first use to reduce provider import time.
Patch Changes
Defer validation schema construction until first use to reduce provider import time.
2.0.86
19 Oct 2026openai-compatible, groq, alibaba: Put reasoning before text in doGenerate
2.0.85
16 Oct 2026openai-compatible: Support URL-based image responses
Add opt-in URL validation and credential origin handling to
getFromApifor URL-based image responses. Validate redirect targets, strip unsafe request headers, and drop credentials on cross-origin redirects while preserving existing unvalidated requests.
Patch Changes
openai-compatible: Support URL-based image responses
Add opt-in URL validation and credential origin handling to
getFromApifor URL-based image responses. Validate redirect targets, strip unsafe request headers, and drop credentials on cross-origin redirects while preserving existing unvalidated requests.
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.