Repository navigation
chore(deps): update dependency ai to v7 - #1907
renovate[bot] wants to merge 1 commit into
Conversation
|
Important Review skippedAuto reviews are limited based on label configuration. 🏷️ Required labels (at least one) (1)
Please check the settings in the CodeRabbit UI or the ⚙️ Run configuration
You can disable this status message by setting the Use the checkbox below for a quick retry:
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
Review statusThis PR was opened by an automated account. A human maintainer must verify the change intent, provenance, and validation before merging. Current step: Awaiting fresh human maintainer or CODEOWNER approval. Review-state labels are managed by this workflow; do not edit them manually. |
Codecov Report✅ All modified and coverable lines are covered by tests. 📢 Thoughts on this report? Let us know! |
github-actions
Bot
added
the
awaiting-maintainer
ad879e9 to
de416c5
Compare
github-actions
Bot
added
awaiting-maintainer
de416c5 to
f63b351
Compare
github-actions
Bot
added
awaiting-maintainer
f63b351 to
06d0d62
Compare
github-actions
Bot
added
awaiting-maintainer
This PR contains the following updates:
6.0.218→7.0.130Release Notes
vercel/ai (ai)
v7.0.130Compare Source
Patch Changes
6f6b9c0: fix(ai): reject pending result promises and cancel the pending stream when the abort signal firesd3bcad9: feat(provider): report decision refusals as refusal answersd3bcad9]v7.0.129Compare Source
Patch Changes
e37d213: feat(ai): emit deprecation warnings with stable codes and deduplicate them73d8343: fix(ai): preserve V2 provider-reported total tokens in raw usage268225c: feat(provider): add a portablemaxreasoning level with native provider mappings, compatibility coercions, and budget-based fallback support.Compatibility notes:
maxwidensLanguageModelV4CallOptions['reasoning']and the derivedReasoningLeveltype. Third-party providers and consumers with exhaustive switches orRecord<ReasoningLevel, ...>mappings must handlemaxbefore upgrading. Ordinary calls using existing reasoning levels remain accepted.reasoning: 'xhigh'now sendmaxReasoningEffort: 'high'instead of the unsupported'max', and return a compatibility warning. Anthropic and OpenAI Bedrock effort mappings retain their existingxhighbehavior. Partial Bedrock reasoning configurations continue to preserve explicit values and derive missing fields; explicit effort/budget overrides and disabled thinking no longer emit warnings for portable mappings that are not sent.minimalandxhighcalls keep their request mappings tolowandhigh, respectively, and now return compatibility warnings in generation and streaming results. Portablemaxmaps tohighwith the same warning behavior.Release classification: This remains a patch changeset under the repository's explicit release policy, which uses patch releases for both fixes and features. The TypeScript source-compatibility caveat for exhaustive consumers is disclosed above. Maintainers can override this classification with a
majorlabel if they decide to align the provider-spec addition with a future AI SDK major release.d83366e: fix(ai): publish the latest error after repeated failed stream resume attempts34905a2: fix(ai): preserve provider options from approval-only tool messages16ab882: fix(ai): recognize WebKit network errorsf810ea3: fix(ai): prevent unavailable tools from exposing persisted output to models19a127b: fix(ai): avoid duplicate slashes in chat reconnect URLs0af2f7c: chore: upgrade to TypeScript 64c6979c: fix(harness): preserve adapter warnings in agent stream results and declaretheir compatibility warning shapes
Updated dependencies [
fc1e19e]Updated dependencies [
5094ebd]Updated dependencies [
753f2e1]Updated dependencies [
268225c]Updated dependencies [
0af2f7c]Updated dependencies [
ad64697]Updated dependencies [
0c82824]v7.0.128Compare Source
Patch Changes
0fe8c67: fix: preserve tool approval state when resuming streams2959d35: feat: renameevaluatetodecide0ca1ab9: Add durable non-streamingWorkflowAgent.generate()with a shared tool loop, a 20-step default, typed output, and core generation result semantics. Reuse core result and content construction through internal exports while preserving existing Workflow streaming behavior. Transport-independent approval creation remains separate follow-up work.d6b42fd: feat(ai): support custom reasoning delimiters in extractReasoningMiddlewareed6e72d: fix(ai): return successful empty transcripts for silent audio2136151: ThrowInvalidResponseDataErrorinstead of a genericErrorwhen a generated audio file's format cannot be determined from its media type, so callers can identify the failure withAISDKError.isInstance.131532b]2959d35]59116e6]6ac923a]v7.0.127Compare Source
Patch Changes
158a718: feat(ai): select and rank eligible deferred tools via 'search()' callback in tool searchbb8d33e: fix(ai): cancel merged UI message streams when the consumer disconnects284dc11: fix(ai): accept unchanged tool approval inputs created in another JavaScript realmba8afe9: feat(ai): add a configurable maxResults for number of tools returned in tool searchd1bb9e8]v7.0.126Compare Source
Patch Changes
4f3d236: fix: clear tool approvals whenaddToolOutputrunsv7.0.124Compare Source
Patch Changes
8c65988: feat(ai): add telemetry to speech generation and transcription, includingprovider usage propagation and experimental streaming lifecycle callbacks
8c65988]e25994e]527a163]e74afc5]v7.0.123Compare Source
Patch Changes
05cdac6: fix(ai): keep idle UI message streams open with optional SSE heartbeats4514fc1: fix(ai): prune all tool content when retaining zero trailing messagesede5b89: chore: migrate package builds from tsup to tsdown2a625cf: fix(ai): preserve partial reasoning tags when streamed text parts end50a26d5: fix(ai): ignore pending tool approvals superseded by user messages040033b]ede5b89]v7.0.122Compare Source
Patch Changes
f3575f8: Clarify that provider-executed tool execution errors bypass the UI stream'sonErrorcallback to preserve provider error data and harness runtime messages. Stream errors and invalid tool calls still use the callback. Runtime behavior is unchanged.27ab8d4: Encode chat IDs in default stream reconnection URLs so slashes, query delimiters, and fragments stay within the ID. Reject standalone.and..IDs before fetching. Custom URLs returned byprepareReconnectToStreamRequestremain unchanged.870f509]a75f1fd]v7.0.120Compare Source
Patch Changes
b032d70: fix(ai): preserve tool metadata from tool output chunkse21b98b: fix(ai): continue active UI message parts when resuming after a disconnecte361d39]v7.0.119Compare Source
Patch Changes
33e94ba: fix(ai): cancel response streams when clients disconnect34d869e: fix(ai): prevent unhandled stream completion rejections in non-Node runtimesdef4df8: fix(ai): store static tool input errors in the current input field525efc5: feat(provider): advertise image model file and mask input supportUse confirmed model IDs for capability declarations so unrecognized model names
remain unknown. Include Together AI FLUX.2 Pro and Flex single-image editing, and
allow asynchronous capability lookups and middleware overrides to resolve to
unknown.
Advertise QuiverAI Arrow 2 and Arrow 2 Telos file-input support, and mark Together
AI Gemini image inputs unsupported by the current single-image request mapping.
Updated dependencies [
e3605f6]Updated dependencies [
525efc5]v7.0.118Compare Source
Patch Changes
e6a7996]e19f0fc]v7.0.117Compare Source
Patch Changes
b67b1b7]v7.0.116Compare Source
Patch Changes
af9597b: Compile packages for ES2022 runtime targetbc49f78: Preserve original opaque URI strings in tagged file URLs during prompt conversion. Match explicit supported URL MIME types exactly so unsupported subtypes are not forwarded to providers.af9597b]bc49f78]v7.0.114Compare Source
Patch Changes
b5679a7: fix(ai): filter tracing-channel context with telemetry allowlistsca31b89: Clarify in the prompt validation code thatallowSystemInMessagespermits all system messages, including instruction text.124b6ef]v7.0.113Compare Source
Patch Changes
dcdb011: fix(ai): route completed streamed tool input callbacks to repaired tools8f72832: fix(ai): preserve video models from legacy fallback providersfe07867: Fix GoogleembedManycalls with more than 100 values by keeping per-value multimodal content aligned across automatic batches, including text-only entries. Validate content length before sending requests and validate each batch's provider options after middleware transforms them.b74c0cb: fix(ai): resume tool approvals from earlier messagesa4b0940: fix(ai): execute manually approved tool inputs produced by schema transformsPreserve approved inputs during revalidation and reject histories whose reconstructed schema output differs, including signed approvals with missing original input. Validate transformed UI tool inputs against the reconstructed output before returning them as static tool parts.
2693319: Add Gemini 3.8 TTS support with structured speech metadata and per-turn speaker and style controls for prebuilt voices. Preserve native WAV responses without adding a second header, support explicit raw PCM, mu-law, and A-law output, and identify headerless audio formats correctly. Add the Gemini 3.8 speech model IDs to Google and Gateway types.Share transcript and custom-voice inspection through the Google provider internal export, and reject empty speech transcripts before sending a request. Default newer and custom model IDs to structured speech while preserving the legacy format for Gemini 2.5 and 3.1.
c93ee90: fix(ai): preserve message history for direct transport stream callbacks771e74b: chore: enable dead code lint rulesUpdated dependencies [
fe07867]Updated dependencies [
a4b0940]Updated dependencies [
2693319]Updated dependencies [
b73f2f9]Updated dependencies [
771e74b]v7.0.112Compare Source
Patch Changes
9a98fd9: fix(ai): prune reasoning file parts when removing reasoninga0553d6: feat(ai): report consumer cancellation in UI message stream end callbacksffb0e76: fix(provider): preserve opaque file URI strings for provider serializationfde0d66: fix(ai): preserve parsed metadata and data values when validating UI messagesed5a1d7]ffb0e76]618dc11]v7.0.111Compare Source
Patch Changes
31d24ce: feat: add telemetry support toexperimental_evaluatea65bfd9: fix(ai): securely download URL-backed language model file outputs so generated files expose actual base64 and byte content and UI streams contain valid data URLs. Propagate cancellation to batch file downloads.v7.0.110Compare Source
Patch Changes
49295bb]v7.0.109Compare Source
Patch Changes
0343bb1: fix(ai): keep replacement completion requests loading and cancellable when an earlier request settles2b105fa: fix(ai): preserve overlapping text blocks in reasoning extraction streams125f493: fix(harness): forward validatedtoolsContextto host-executed tools in alignment withToolLoopAgentv7.0.108Compare Source
Patch Changes
3f6852a: fix(ai): prevent direct execution of tools governed by tool callers6317504: fix(ai): stop pending tool-call repairs when generation is cancelled3cb2dcd: fix(ai): preserve file data when adapting v3 language modelsccf98e7: fix(ai): preventstreamTextfrom executing tool calls that violate tool choice20dd00a]7cf7cee]c42576a]d85dcf5]fd9b3f3]v7.0.107Compare Source
Patch Changes
79681c4: fix(ai): preserve provider file and skill upload APIs in wrapProvider98c7275: fix(ai): preserve query parameters in chat reconnect URLsa105059: fix(workflow): support deferred tool discovery in WorkflowAgent31532f3: fix(ai): prevent preliminary tool outputs from completing chatse61cbd8: fix(ai): preserve raw speech audio format metadata8ade040: fix(ai): pass tool-specific context to input callbacks970a01e: fix(ai): enforce polling timeouts for in-flight video status requests85539c5: fix(ai): preserve multiple Set-Cookie headers in Node stream responses611d301: fix(ai): prevent duplicate content types in chat transport requestsc415657: fix(ai): decode base64 text data URLs using their declared charset2973485]a4db5ea]2937ea2]v7.0.106Compare Source
Patch Changes
4775577: fix(ai): preserve provider metadata when simulating text streams6696728: fix(ai): report the prepareStep model in streamed step results09516a1: fix(ai): prevent unhandled rejections when UI message stream reading stops early1aef01e: fix(ai): preserve prototype-named properties in serialized tool outputs9c1ea74: fix(ai): close telemetry spans when provider response streams fail107343a: fix(ai): use the prepareStep-selected model for streamed response metadata fallbacks03c3e33: fix(ai): preserve tool calls required by retained pending approvals5d42ebd: fix(ai): skip input available callbacks for invalid streamed tool calls4a67783: fix(ai): cancel prompt attachment downloads when model calls are aborted or time out1058ed5: fix(ai): strip streamed JSON fences before arbitrary trailing whitespace84f5d1b: fix(ai): stream null and empty string JSON partial outputs2d53a5d: fix(ai): prevent onEnd after aborting a multi-step text stream2a5ed55: fix(ai): stream structured output from the final tool-loop step4fdf51e]0455398]v7.0.105Compare Source
Patch Changes
6982e9d: Resolve evaluation model IDs through AI Gateway when no default provider is configured, including string aliases in custom providers.6982e9d]6982e9d]v7.0.104Compare Source
Patch Changes
a7dd893: Add experimental evaluation model aliases and registry resolution.customProvideracceptsevaluationModels, registries exposeevaluationModel, andexperimental_evaluateaccepts string IDs when an evaluation-capable default provider is explicitly configured. Evaluation never implicitly falls back to Gateway. Model-resolution errors now identifyevaluationModelwhile stable provider contracts remain unchanged.227f3b0: fix(ai): report abnormal realtime WebSocket close diagnostics throughonError3456e2c: feat(ai): support tool search with direct tool callingc4e76de: feat(ai): add native tool search tool215b25e]d4d96bf]a7dd893]3456e2c]c4e76de]v7.0.103Compare Source
Patch Changes
91c2128: feat(ai): add mid conversation tool discovery/updates for code-mode25a0447: feat(ai): deprecate rawInput in output-error UI message parts2cd80b3: Keep default Node.js downloads protected by DNS validation and connection pinning when frameworks or instrumentation wrap global fetch before or after the SDK loads.123d71f: Addexperimental_evaluateand the isolated experimental v4 evaluation model specification for Choice, Score, and Boolean questions against shared state. Includes typed answers, optional Choice/Score distributions, required Boolean probabilities, validation, retries, cancellation, andExperimental_EvaluationUnsupportedQuestionTypeErrorfor unsupported questions.91c2128]0c9ab5a]2cd80b3]d06bb2a]123d71f]2fa5e0e]2cce7da]v7.0.102Compare Source
Patch Changes
5c0054d: Add optional browser-direct WebRTC for experimental client-delegated Live conversations alongside the existing WebSocket path. Exchange SDP through an application endpoint withapi.session, configure server-owned data-channel permissions, and preserve committed React session ownership. Capture follows the selected sender track, borrowed tracks remain caller-owned, and disconnect recovery and finalization stay bounded. Applications continue to handle client delegation and submit context; Live session updates and Responses delegation remain unsupported.Serialize microphone sender changes and close the peer if detachment fails, without stopping borrowed tracks. Validate nonempty SDP setup answers with a bounded response body, and document the same-origin broker authentication contract.
39535af: Add experimental OpenAI Live provider support through the unifiedopenai.experimental_realtimefactory for server WebSocket sessions with client delegation. Applications own their agents and tools, receive continuous audio/transcript events and delegation metadata, and return context through validated channels. Support immutable startup options, microphone mute controls, graceful session close, and cumulative voice usage. Responses delegation and Live session updates reject before sending.Route known Live model IDs to Live, allow an OpenAI-specific
apioverride for early-access models, and preserve legacy Realtime defaults for unknown IDs. Token minting follows the same selection rules and rejects Live before requesting unsupported credentials. Extend the realtime v4 specification with optional server WebSocket configuration, per-connection raw-event parsers, and model-wide startup/finalization capabilities. This provider layer supplies connection settings and protocol mapping for server adapters; browser lifecycle and UI integration belong to the core runtime and framework hooks.Preserve Realtime client event IDs for session updates and audio appends, and correlate server errors with the originating client event.
8b92ba9: fix(ai): settle automatically denied tool calls in UI streams4b306c2: Report abnormal realtime WebSocket close codes and reasons through the session error path.4b306c2: Add the client-delegation-first realtime WebSocket runtime with session lifecycle, exact final duration, bounded event queues and command acknowledgement tracking, transcripts, context append, and reversible capture. Continuous sessions support PCM16 audio-chunk playback over an application relay with a recoverable live-edge buffer policy. Legacy turn-based WebSockets retain queued playback and frontend tool handling.Remove the deferred managed Responses coordinator, autoContinueTools option, backend usage state, and Live tool-result aliases. Continuous text and delegation handling belong to the application; sendTextMessage and addToolOutput reject continuous sessions. Server-confirmed provider delegation and turn-based audio-delta on the continuous profile fail explicitly. maxPlaybackBufferSeconds is supported only for continuous PCM sessions.
Fence callbacks, startup publications, tool results, and teardown by connection attempt. Validate token setup before opening a client-secret socket, require explicit relay transport mode, and include WebSocket URL, protocol values, and runtime timeouts/buffer settings in React session configuration keys.
Retain immediate local capture for explicit legacy preconnect streams, including owned-track cleanup and capture continuity through asynchronous setup. Signal transport closing before draining accepted terminal events, share reentrant close settlement, and cancel stale readiness and deadlines. Enforce 128 KiB frame and combined buffered-send budgets only for Live continuous sessions, using actual encoded wire bytes; oversized manual operations remain recoverable while startup and automatic audio failures close the session. Legacy startup, text, tool output, and audio retain their pre-Live behavior without these byte caps. Live pending-audio overflow drains accepted terminal usage within the existing one-second drain window. Validate final provider-transformed WebSocket URL syntax without removing native authentication query parameters.
Updated dependencies [
5c0054d]Updated dependencies [
39535af]v7.0.101Compare Source
Patch Changes
6aa7c54: fix(ai): serialize tool output JSON valuesv7.0.100Compare Source
Patch Changes
6431635: feat(ai): expose typed AI SDK errors for UI transport and completion failures23a0fff]v7.0.99Compare Source
Patch Changes
615ac89: feat(ai): use InvalidArgumentError for utility input validation7f76d83]v7.0.98Patch Changes
5ec21a6: fix: reject unsupported batch request typesdb59d78: feat(ai): add runtime context attribution to embed, embedMany and rerank7469a3b: feat: support image generation requests in batchesf87bf07: fix(ai): reject invalid reranking provider indicesbc5cb7a: fix(ai): accept inferred tools invalidateUIMessagesa5f449a: feat(ai): add a stable UI message type and type guard for tool output errors5ec21a6]7469a3b]dbd83a3]813bb36]c43e4b7]03f4e59]v7.0.97Patch Changes
ef3bac4: Observe video webhook receiver rejections before generation starts to prevent unhandled rejections during or after a failed start. Preserve start error precedence and assimilate custom receivers only once.9942196: feat: add batch cancel and list APIs9942196]v7.0.96Patch Changes
912fb01: feat: add batch cancel and list APIsc595e6e: fix(ai): callatobwithout a receiver for Cloudflare Workers compatibility912fb01]aa4cc14]f102e41]v7.0.95Compare Source
Patch Changes
27f6d7a: fix(ai): reject embedding model responses that contain no embeddingsv7.0.94Compare Source
Patch Changes
a4ba394: feat: support per-request models in batch36b3364: fix(ai): enforce tool choices in streamText45099da: Retry unclassified empty image results, preserve retry-attempt accounting, add provider-independent result retryability classification, and mark Google and Google Vertex prompt blocks as terminal.a4ba394]e9bf5e3]45099da]56c004c]9e1d1b2]a495511]v7.0.93Compare Source
Patch Changes
df6c009: fix(ai): use new message ID when replacing a message insendMessage6ee74a3: fix(ai): preserve tool part titles when validating UI messagesf13d371: fix(ai): preserve provider metadata when converting failed tool callsd4485fe: feat(ai): support minItems and maxItems in array outputs4f201cc: chore(ai): formally include already supportedonLanguageModelCallStartandonLanguageModelCallEndinToolLoopAgentSettingstype8cdb2a7: fix(ai): decode text data URLs in Node.js0f2281e: fix(ai): reject embedding responses whose count does not match the input valuesfc8e8ac: fix(ai): preserve image call diagnostics when no image is generatedee8391e: fix(ai): support abort signals when the global AbortSignal is not a constructor3cfc1fc]v7.0.92Patch Changes
a51cc94: fix(ai): preserve provider metadata from empty smooth stream deltasd1904d3: fix(ai): surface fallback errors for empty HTTP response bodies84e5a79: fix(ai): skipsmoothStreamdelays while the document is hiddena8e8ad0: fix(ai): expose call ID and abort reason in streamText onAbort callbacksa7e324b]v7.0.91Patch Changes
802af1e: Add configurable recovery for provider errors received afterstreamTextresponse streaming begins. Explicitly configuringstreamRetriesenables isolated retry attempts, including one bounded callback-directed recovery throughStreamTextOnErrorRetryCallbackwithstreamRetries: 0; recovered results and metadata reflect only the successful attempt, while the existingStreamTextOnErrorCallbackcontract and logging-only observer behavior remain compatible.5484f27]36eb7ee]622fa7f]v7.0.90Patch Changes
4d25a08]6bcc0f8]v7.0.89Patch Changes
5190b67: feat(provider): extend the FilesV4 interface with optionalgetFileMetadata,downloadFile(streaming), anddeleteFileoperations, plusabortSignal/headerscall options and a{ type: 'stream' }upload data variant; upload results now exposebyteSize,createdAt, andexpiresAt(also surfaced by the coreuploadFile()helper, which now forwardsabortSignal/headers); addpostMultipartStreamToApi(streaming multipart uploads with deterministic part ordering and failure-path stream teardown),deleteFromApi, andcreateBinaryStreamResponseHandlerto provider-utils5190b67]v7.0.88Patch Changes
8b6b756: fix(ai): prevent generateText from accepting responses that violate required tool choicese07b577: feat: add tool calling support to batchv7.0.87Patch Changes
850d863: fix(ai): preserve approval descriptors in UI message streamsv7.0.86Compare Source
Patch Changes
11109ae: feat(ai): support signed tool approvals in WorkflowAgent1329d5a]v7.0.85Compare Source
Patch Changes
55a9981: Ensure canonical hashes preserve undefined array element positions.dd32de2: fix(ai): sum Gateway image-generation costs across split requestsaa45741: fix(provider/anthropic): preserve native message batch request counts in provider metadata and support the full language-model option surface in batch requestscc29073: feat(ai): expose individual image generation callsd2507af]aa45741]v7.0.84Patch Changes
6669d69: Expose parsed structured output instreamTextend callbacks.a6463ca: fix(ai): allow tool approval secrets in ToolLoopAgent settings and prepareCalle604532: fix(ai): handle stateful and empty-match regular expressions in smoothStream805bbfc]90192f1]v7.0.83Patch Changes
8dd86a9: Validate persisted typed tool calls against current input and output schemas.Schema-incompatible empty or error inputs and terminal history from unavailable
tools remain loadable as dynamic tool parts instead of exposing unvalidated
values under current static tool types.
fda13b3: Allow chats to continue automatically after tool approval denials reach theoutput-deniedstate.957146c: add operation-level outcomes to UI message stream end callbacksce6849a: fix(ai): handle stitchable stream cancellation before an inner stream is registeredv7.0.82Patch Changes
3e125ba: Allow manual tool approval statuses to include a reason and preserve it acrosscore, model, and UI approval requests. OPA
requires-approvaldecisions nowsurface their reason to human approvers. UI request chunks serialize the
optional
reason, while UI messages retain it asapproval.requestReasonseparately from an approver's response
reason.0e7994c]3e125ba]v7.0.79Compare Source
Patch Changes
b251584: Preserve active text and reasoning parts when another merged UI message stream finishes a step, and align workflow stream normalization with the explicit part end chunks.591d25b: feat: add batch completion webhooks.experimental_startTextBatchaccepts awebhookUrl, and the gateway provider registers it through the batchcallbackUrlcontract and exports typed async-job metadata. Direct Anthropic and OpenAI batch providers return an unsupported warning when the option is provided.9de0baf: Parse structured generateText output when providers omit finishReason but return text.591d25b]v7.0.78Compare Source
Patch Changes
96970bb](https://redirect.github.com/vercel/ai/Configuration
📅 Schedule: (in timezone America/Los_Angeles)
🚦 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.
This PR was generated by Mend Renovate. View the repository job log.