# AI SDK: what changed from 6 to 7

- Product: AI SDK (https://whatsnew.fyi/product/ai-sdk)
- Vendor: Vercel
- Range: changelog entries numbered after 6.0.276 up to and including 7.0.92, stable releases only
- Entries below: 41 releases (newest first)
- Resolved: 6 is 6.0.276 and 7 is 7.0.92, the newest stable release of each major we track
- Carrying security changes: 0 · CVEs mentioned: 0 · Mentioning breaking changes: 0 · Removing or deprecating something: 0
- Page: https://whatsnew.fyi/product/ai-sdk/compare/6...7

What's New is an index, not a publisher: every entry below links to the vendor's own release notes, which are the authoritative source. Entries are labelled where they are hand-curated sample data, pre-releases, or drawn from a secondary source such as a developer blog.

Reuse: the summaries, labels and curation here are © What's New. Quote freely with attribution and a link back; wholesale republication of the corpus is not permitted — terms: https://whatsnew.fyi/terms. The vendors' own release notes remain their publishers'.

## What changed (88 changes, grouped by kind)

### Added

#### 7.0.91 (2026-09-02)

- Add configurable recovery for provider errors received after streamText response streaming begins through streamRetries option
- Add StreamTextOnErrorRetryCallback for bounded callback-directed recovery with streamRetries: 0

#### 7.0.89 (2026-09-01)

- Extend the FilesV4 interface with optional getFileMetadata, downloadFile (streaming), and deleteFile operations
- Add abortSignal and headers call options to FilesV4 interface
- Add { type: 'stream' } upload data variant to FilesV4 interface
- Expose byteSize, createdAt, and expiresAt in upload results
- Surface byteSize, createdAt, and expiresAt through the core uploadFile() helper
- Add postMultipartStreamToApi function for streaming multipart uploads with deterministic part ordering and failure-path stream teardown
- Add deleteFromApi function to provider-utils
- Add createBinaryStreamResponseHandler function to provider-utils

#### 7.0.88 (2026-09-01)

- Add tool calling support to batch

#### 7.0.86 (2026-08-31)

- support signed tool approvals in WorkflowAgent

#### 7.0.85 (2026-08-30)

- Expose individual image generation calls

#### 7.0.84 (2026-08-28)

- Expose parsed structured output in streamText end callbacks

#### 7.0.83 (2026-08-26)

- Add operation-level outcomes to UI message stream end callbacks

#### 7.0.79 (2026-08-25)

- Add batch completion webhooks where experimental_startTextBatch accepts a webhookUrl, and the gateway provider registers it through the batch callbackUrl contract and exports typed async-job metadata

#### 7.0.62 (2026-08-11)

- Add typed custom bodies to Completion APIs

#### 7.0.58 (2026-08-07)

- Allow `aspectRatio: 'adaptive'` on `generateVideo` for video models that derive output ratio from input.

#### 7.0.56 (2026-08-07)

- Expose provider metadata on language-model-call end callbacks and telemetry spans

#### 7.0.55 (2026-08-06)

- Add batch APIs

#### 7.0.54 (2026-08-06)

- Allow `ToolLoopAgent` `prepareCall` callbacks to read and override the top-level `reasoning` option
- Add `defaultInstructionsMiddleware` for applying default language model instructions while preserving call-level overrides

#### 7.0.50 (2026-08-03)

- Add asynchronous start/status flow to the experimental video model interface (VideoModelV4) with doStart, doStatus, and handleWebhookOption methods
- Add poll and webhook options to experimental_generateVideo for asynchronous completion orchestration
- Add custom delay implementation support for polling configuration to enable durable workflow compatibility

#### 7.0.45 (2026-07-31)

- Support experimental_toolCallers in streamText
- Support experimental tool callers in ToolLoopAgent

### Changed

#### 7.0.91 (2026-09-02)

- Update @ai-sdk/gateway to 4.0.73

#### 7.0.90 (2026-09-02)

- Updated @ai-sdk/gateway to version 4.0.72
- Updated @ai-sdk/provider-utils to version 5.0.36

#### 7.0.89 (2026-09-01)

- Forward abortSignal and headers in the core uploadFile() helper

#### 7.0.83 (2026-08-26)

- Validate persisted typed tool calls against current input and output schemas, with schema-incompatible inputs loading as dynamic tool parts instead of exposing unvalidated values
- Allow chats to continue automatically after tool approval denials reach the output-denied state

#### 7.0.82 (2026-08-26)

- Allow manual tool approval statuses to include a reason and preserve it across core, model, and UI approval requests
- OPA requires-approval decisions now surface their reason to human approvers
- UI request chunks serialize the optional reason, while UI messages retain it as approval.requestReason separately from an approver's response reason

#### 7.0.79 (2026-08-25)

- Direct Anthropic and OpenAI batch providers return an unsupported warning when the webhookUrl option is provided

#### 7.0.74 (2026-08-21)

- Updated @ai-sdk/gateway to 4.0.60

#### 7.0.73 (2026-08-21)

- Updated @ai-sdk/gateway to version 4.0.59

#### 7.0.69 (2026-08-19)

- Updated @ai-sdk/gateway to version 4.0.55

#### 7.0.68 (2026-08-18)

- Updated @ai-sdk/gateway to version 4.0.54

#### 7.0.67 (2026-08-18)

- Updated @ai-sdk/gateway to version 4.0.53

#### 7.0.64 (2026-08-12)

- @ai-sdk/gateway dependency updated to version 4.0.51

#### 7.0.63 (2026-08-12)

- Allow providers without reranking model support to be assigned to Provider

#### 7.0.58 (2026-08-07)

- Reduce bundle size by making internal Zod v4 imports tree-shakeable.

#### 7.0.57 (2026-08-07)

- Updated @ai-sdk/provider-utils to 5.0.24
- Updated @ai-sdk/gateway to 4.0.45

#### 7.0.50 (2026-08-03)

- Simplify tool caller configuration in code-mode

#### 7.0.49 (2026-08-03)

- Updated @ai-sdk/gateway to version 4.0.38

#### 7.0.48 (2026-08-01)

- @ai-sdk/gateway updated to version 4.0.37

#### 7.0.47 (2026-07-31)

- Updated @ai-sdk/provider-utils to version 5.0.18
- Updated @ai-sdk/gateway to version 4.0.36

#### 7.0.46 (2026-07-31)

- Updated @ai-sdk/gateway to version 4.0.35

### Fixed

#### 7.0.92 (2026-09-03)

- preserve provider metadata from empty smooth stream deltas
- surface fallback errors for empty HTTP response bodies
- skip smoothStream delays while the document is hidden
- expose call ID and abort reason in streamText onAbort callbacks

#### 7.0.88 (2026-09-01)

- Prevent generateText from accepting responses that violate required tool choices

#### 7.0.87 (2026-08-31)

- Preserve approval descriptors in UI message streams

#### 7.0.85 (2026-08-30)

- Ensure canonical hashes preserve undefined array element positions
- Sum Gateway image-generation costs across split requests
- Preserve native message batch request counts in provider metadata and support the full language-model option surface in batch requests

#### 7.0.84 (2026-08-28)

- Allow tool approval secrets in ToolLoopAgent settings and prepareCall
- Handle stateful and empty-match regular expressions in smoothStream

#### 7.0.83 (2026-08-26)

- Handle stitchable stream cancellation before an inner stream is registered

#### 7.0.79 (2026-08-25)

- 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
- Parse structured generateText output when providers omit finishReason but return text

#### 7.0.76 (2026-08-21)

- Prevent duplicate text and reasoning part ids
- Allow nullish metadata schemas for UI messages with branded IDs

#### 7.0.71 (2026-08-20)

- Prevent exceptions in streaming onChunk and onError callbacks from terminating the stream or masking provider errors

#### 7.0.70 (2026-08-19)

- Stop multi-step text generation for client tool approval even when a provider-executed tool has a deferred result
- Reject streamObject result promises and report failed completion when the provider stream errors
- Filter preliminary tool outputs when ignoreIncompleteToolCalls is enabled
- Prevent automatic tool execution when a model call ends with an unsafe finish reason
- Clear partial UI message parts when a WorkflowAgent model-call step is retried

#### 7.0.66 (2026-08-14)

- Keep chat status submitted until response content begins streaming
- Fix declaration emit for exported values that infer an Output type

#### 7.0.65 (2026-08-13)

- Avoid repeatedly cloning accumulated text in readUIMessageStream while preserving independent snapshots for mutable nested values
- Preserve root-level JSON Schema definitions when wrapping array output schemas
- Propagate errors thrown by the Chat onFinish callback to the initiating request

#### 7.0.63 (2026-08-12)

- Preserve reasoning block IDs from UI message streams on reasoning UI parts

#### 7.0.58 (2026-08-07)

- Respect ToolLoopAgent timeouts configured in agent settings.

#### 7.0.56 (2026-08-07)

- Make retried doStart calls idempotent by minting one idempotency token per logical start and forwarding it as an idempotency-key header to prevent duplicate billable generations on retry after lost response
- Stop pending and active resumed chat streams after cancellation and prevent overlapping resumptions from applying stale updates

#### 7.0.54 (2026-08-06)

- Preserve preceding assistant messages when regenerating a response

#### 7.0.52 (2026-08-04)

- Skip re-validating tool input for terminal output-available UI message parts

#### 7.0.45 (2026-07-31)

- Warn when generateText receives streaming-only firstChunkMs or chunkMs timeout settings

#### 7.0.44 (2026-07-31)

- Fix telemetry attribution for language model calls that resolve to a different response model

_3 releases carry no categorized changes yet: 7.0.77, 7.0.59, 7.0.51._

## Release notes

### 7.0.92

- Date: 2026-09-03
- Version: 7.0.92
- Original notes: https://github.com/vercel/ai/releases/tag/ai%407.0.92
- Permalink: https://whatsnew.fyi/product/ai-sdk/releases/7.0.92

- **fixed** — preserve provider metadata from empty smooth stream deltas
- **fixed** — surface fallback errors for empty HTTP response bodies
- **fixed** — skip smoothStream delays while the document is hidden
- **fixed** — expose call ID and abort reason in streamText onAbort callbacks

###### Patch Changes

- a51cc94: fix(ai): preserve provider metadata from empty smooth stream deltas
- d1904d3: fix(ai): surface fallback errors for empty HTTP response bodies
- 84e5a79: fix(ai): skip `smoothStream` delays while the document is hidden
- a8e8ad0: fix(ai): expose call ID and abort reason in streamText onAbort callbacks
- Updated dependencies [a7e324b]
  - @ai-sdk/gateway@4.0.74

### 7.0.91

- Date: 2026-09-02
- Version: 7.0.91
- Original notes: https://github.com/vercel/ai/releases/tag/ai%407.0.91
- Permalink: https://whatsnew.fyi/product/ai-sdk/releases/7.0.91

- **added** — Add configurable recovery for provider errors received after streamText response streaming begins through streamRetries option
- **added** — Add StreamTextOnErrorRetryCallback for bounded callback-directed recovery with streamRetries: 0
- **changed** — Update @ai-sdk/gateway to 4.0.73

###### Patch Changes

- 802af1e: Add configurable recovery for provider errors received after `streamText` response streaming begins. Explicitly configuring `streamRetries` enables isolated retry attempts, including one bounded callback-directed recovery through `StreamTextOnErrorRetryCallback` with `streamRetries: 0`; recovered results and metadata reflect only the successful attempt, while the existing `StreamTextOnErrorCallback` contract and logging-only observer behavior remain compatible.
- Updated dependencies [5484f27]
- Updated dependencies [36eb7ee]
- Updated dependencies [622fa7f]
  - @ai-sdk/gateway@4.0.73

### 7.0.90

- Date: 2026-09-02
- Version: 7.0.90
- Original notes: https://github.com/vercel/ai/releases/tag/ai%407.0.90
- Permalink: https://whatsnew.fyi/product/ai-sdk/releases/7.0.90

- **changed** — Updated @ai-sdk/gateway to version 4.0.72
- **changed** — Updated @ai-sdk/provider-utils to version 5.0.36

###### Patch Changes

- Updated dependencies [4d25a08]
- Updated dependencies [6bcc0f8]
  - @ai-sdk/gateway@4.0.72
  - @ai-sdk/provider-utils@5.0.36

### 7.0.89

- Date: 2026-09-01
- Version: 7.0.89
- Original notes: https://github.com/vercel/ai/releases/tag/ai%407.0.89
- Permalink: https://whatsnew.fyi/product/ai-sdk/releases/7.0.89

- **added** — Extend the FilesV4 interface with optional getFileMetadata, downloadFile (streaming), and deleteFile operations
- **added** — Add abortSignal and headers call options to FilesV4 interface
- **added** — Add { type: 'stream' } upload data variant to FilesV4 interface
- **added** — Expose byteSize, createdAt, and expiresAt in upload results
- **added** — Surface byteSize, createdAt, and expiresAt through the core uploadFile() helper
- **changed** — Forward abortSignal and headers in the core uploadFile() helper
- **added** — Add postMultipartStreamToApi function for streaming multipart uploads with deterministic part ordering and failure-path stream teardown
- **added** — Add deleteFromApi function to provider-utils
- **added** — Add createBinaryStreamResponseHandler function to provider-utils

###### Patch Changes

- 5190b67: feat(provider): extend the FilesV4 interface with optional `getFileMetadata`, `downloadFile` (streaming), and `deleteFile` operations, plus `abortSignal`/`headers` call options and a `{ type: 'stream' }` upload data variant; upload results now expose `byteSize`, `createdAt`, and `expiresAt` (also surfaced by the core `uploadFile()` helper, which now forwards `abortSignal`/`headers`); add `postMultipartStreamToApi` (streaming multipart uploads with deterministic part ordering and failure-path stream teardown), `deleteFromApi`, and `createBinaryStreamResponseHandler` to provider-utils
- Updated dependencies [5190b67]
  - @ai-sdk/provider@4.0.10
  - @ai-sdk/provider-utils@5.0.35
  - @ai-sdk/gateway@4.0.71

### 7.0.88

- Date: 2026-09-01
- Version: 7.0.88
- Original notes: https://github.com/vercel/ai/releases/tag/ai%407.0.88
- Permalink: https://whatsnew.fyi/product/ai-sdk/releases/7.0.88

- **fixed** — Prevent generateText from accepting responses that violate required tool choices
- **added** — Add tool calling support to batch

###### Patch Changes

- 8b6b756: fix(ai): prevent generateText from accepting responses that violate required tool choices
- e07b577: feat: add tool calling support to batch

### 7.0.87

- Date: 2026-08-31
- Version: 7.0.87
- Original notes: https://github.com/vercel/ai/releases/tag/ai%407.0.87
- Permalink: https://whatsnew.fyi/product/ai-sdk/releases/7.0.87

- **fixed** — Preserve approval descriptors in UI message streams

###### Patch Changes

- 850d863: fix(ai): preserve approval descriptors in UI message streams

### 7.0.86

- Date: 2026-08-31
- Version: 7.0.86
- Original notes: https://github.com/vercel/ai/releases/tag/ai%407.0.86
- Permalink: https://whatsnew.fyi/product/ai-sdk/releases/7.0.86

- **added** — support signed tool approvals in WorkflowAgent

###### Patch Changes

- 11109ae: feat(ai): support signed tool approvals in WorkflowAgent
- Updated dependencies [1329d5a]
  - @ai-sdk/gateway@4.0.70

### 7.0.85

- Date: 2026-08-30
- Version: 7.0.85
- Original notes: https://github.com/vercel/ai/releases/tag/ai%407.0.85
- Permalink: https://whatsnew.fyi/product/ai-sdk/releases/7.0.85

- **fixed** — Ensure canonical hashes preserve undefined array element positions
- **fixed** — Sum Gateway image-generation costs across split requests
- **fixed** — Preserve native message batch request counts in provider metadata and support the full language-model option surface in batch requests
- **added** — Expose individual image generation calls

###### Patch Changes

- 55a9981: Ensure canonical hashes preserve undefined array element positions.
- dd32de2: fix(ai): sum Gateway image-generation costs across split requests
- aa45741: fix(provider/anthropic): preserve native message batch request counts in provider metadata and support the full language-model option surface in batch requests
- cc29073: feat(ai): expose individual image generation calls
- Updated dependencies [d2507af]
- Updated dependencies [aa45741]
  - @ai-sdk/gateway@4.0.69
  - @ai-sdk/provider@4.0.9
  - @ai-sdk/provider-utils@5.0.34

### 7.0.84

- Date: 2026-08-28
- Version: 7.0.84
- Original notes: https://github.com/vercel/ai/releases/tag/ai%407.0.84
- Permalink: https://whatsnew.fyi/product/ai-sdk/releases/7.0.84

- **added** — Expose parsed structured output in streamText end callbacks
- **fixed** — Allow tool approval secrets in ToolLoopAgent settings and prepareCall
- **fixed** — Handle stateful and empty-match regular expressions in smoothStream

###### Patch Changes

- 6669d69: Expose parsed structured output in `streamText` end callbacks.
- a6463ca: fix(ai): allow tool approval secrets in ToolLoopAgent settings and prepareCall
- e604532: fix(ai): handle stateful and empty-match regular expressions in smoothStream
- Updated dependencies [805bbfc]
- Updated dependencies [90192f1]
  - @ai-sdk/gateway@4.0.68
  - @ai-sdk/provider-utils@5.0.33

### 7.0.83

- Date: 2026-08-26
- Version: 7.0.83
- Original notes: https://github.com/vercel/ai/releases/tag/ai%407.0.83
- Permalink: https://whatsnew.fyi/product/ai-sdk/releases/7.0.83

- **changed** — Validate persisted typed tool calls against current input and output schemas, with schema-incompatible inputs loading as dynamic tool parts instead of exposing unvalidated values
- **changed** — Allow chats to continue automatically after tool approval denials reach the output-denied state
- **added** — Add operation-level outcomes to UI message stream end callbacks
- **fixed** — Handle stitchable stream cancellation before an inner stream is registered

###### Patch 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 the `output-denied` state.
- 957146c: add operation-level outcomes to UI message stream end callbacks
- ce6849a: fix(ai): handle stitchable stream cancellation before an inner stream is registered

### 7.0.82

- Date: 2026-08-26
- Version: 7.0.82
- Original notes: https://github.com/vercel/ai/releases/tag/ai%407.0.82
- Permalink: https://whatsnew.fyi/product/ai-sdk/releases/7.0.82

- **changed** — Allow manual tool approval statuses to include a reason and preserve it across core, model, and UI approval requests
- **changed** — OPA requires-approval decisions now surface their reason to human approvers
- **changed** — UI request chunks serialize the optional reason, while UI messages retain it as approval.requestReason separately from an approver's response reason

###### Patch Changes

- 3e125ba: Allow manual tool approval statuses to include a reason and preserve it across
  core, model, and UI approval requests. OPA `requires-approval` decisions now
  surface their reason to human approvers. UI request chunks serialize the
  optional `reason`, while UI messages retain it as `approval.requestReason`
  separately from an approver's response `reason`.
- Updated dependencies [0e7994c]
- Updated dependencies [3e125ba]
  - @ai-sdk/gateway@4.0.67
  - @ai-sdk/provider-utils@5.0.32

### 7.0.79

- Date: 2026-08-25
- Version: 7.0.79
- Original notes: https://github.com/vercel/ai/releases/tag/ai%407.0.79
- Permalink: https://whatsnew.fyi/product/ai-sdk/releases/7.0.79

- **fixed** — 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
- **added** — Add batch completion webhooks where experimental_startTextBatch accepts a webhookUrl, and the gateway provider registers it through the batch callbackUrl contract and exports typed async-job metadata
- **changed** — Direct Anthropic and OpenAI batch providers return an unsupported warning when the webhookUrl option is provided
- **fixed** — Parse structured generateText output when providers omit finishReason but return text

###### 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_startTextBatch` accepts a `webhookUrl`, and the gateway provider registers it through the batch `callbackUrl` contract 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.
- Updated dependencies [591d25b]
  - @ai-sdk/gateway@4.0.64
  - @ai-sdk/provider@4.0.8
  - @ai-sdk/provider-utils@5.0.30

### 7.0.77

- Date: 2026-08-22
- Version: 7.0.77
- Original notes: https://github.com/vercel/ai/releases/tag/ai%407.0.77
- Permalink: https://whatsnew.fyi/product/ai-sdk/releases/7.0.77

###### Patch Changes

-   Updated dependencies [b74971f]
-   Updated dependencies [a371615]
    -   @ai-sdk/provider-utils@5.0.29
    -   @ai-sdk/gateway@4.0.62

### 7.0.76

- Date: 2026-08-21
- Version: 7.0.76
- Original notes: https://github.com/vercel/ai/releases/tag/ai%407.0.76
- Permalink: https://whatsnew.fyi/product/ai-sdk/releases/7.0.76

- **fixed** — Prevent duplicate text and reasoning part ids
- **fixed** — Allow nullish metadata schemas for UI messages with branded IDs

###### Patch Changes

-   c6d57f3: fix(ai): prevent duplicate text and reasoning part ids
-   677a707: fix(ai): allow nullish metadata schemas for UI messages with branded IDs

### 7.0.74

- Date: 2026-08-21
- Version: 7.0.74
- Original notes: https://github.com/vercel/ai/releases/tag/ai%407.0.74
- Permalink: https://whatsnew.fyi/product/ai-sdk/releases/7.0.74

- **changed** — Updated @ai-sdk/gateway to 4.0.60

###### Patch Changes

-   Updated dependencies [5941bd6]
-   Updated dependencies [9a4337d]
    -   @ai-sdk/gateway@4.0.60

### 7.0.73

- Date: 2026-08-21
- Version: 7.0.73
- Original notes: https://github.com/vercel/ai/releases/tag/ai%407.0.73
- Permalink: https://whatsnew.fyi/product/ai-sdk/releases/7.0.73

- **changed** — Updated @ai-sdk/gateway to version 4.0.59

###### Patch Changes

-   Updated dependencies [f607a12]
    -   @ai-sdk/gateway@4.0.59

### 7.0.71

- Date: 2026-08-20
- Version: 7.0.71
- Original notes: https://github.com/vercel/ai/releases/tag/ai%407.0.71
- Permalink: https://whatsnew.fyi/product/ai-sdk/releases/7.0.71

- **fixed** — Prevent exceptions in streaming onChunk and onError callbacks from terminating the stream or masking provider errors

###### Patch Changes

-   9a37469: Prevent exceptions in streaming `onChunk` and `onError` callbacks from terminating the stream or masking provider errors.
-   Updated dependencies [936719b]
    -   @ai-sdk/gateway@4.0.57

### 7.0.70

- Date: 2026-08-19
- Version: 7.0.70
- Original notes: https://github.com/vercel/ai/releases/tag/ai%407.0.70
- Permalink: https://whatsnew.fyi/product/ai-sdk/releases/7.0.70

- **fixed** — Stop multi-step text generation for client tool approval even when a provider-executed tool has a deferred result
- **fixed** — Reject streamObject result promises and report failed completion when the provider stream errors
- **fixed** — Filter preliminary tool outputs when ignoreIncompleteToolCalls is enabled
- **fixed** — Prevent automatic tool execution when a model call ends with an unsafe finish reason
- **fixed** — Clear partial UI message parts when a WorkflowAgent model-call step is retried

###### Patch Changes

-   9566914: Stop multi-step text generation for client tool approval even when a provider-executed tool has a deferred result.
-   b181020: fix: reject `streamObject` result promises and report failed completion when the provider stream errors
-   7054073: Filter preliminary tool outputs when `ignoreIncompleteToolCalls` is enabled.
-   a828527: Prevent automatic tool execution when a model call ends with an unsafe finish reason.
-   d3cc3fe: Clear partial UI message parts when a WorkflowAgent model-call step is retried.
-   Updated dependencies [e6087c9]
    -   @ai-sdk/provider-utils@5.0.28
    -   @ai-sdk/gateway@4.0.56

### 7.0.69

- Date: 2026-08-19
- Version: 7.0.69
- Original notes: https://github.com/vercel/ai/releases/tag/ai%407.0.69
- Permalink: https://whatsnew.fyi/product/ai-sdk/releases/7.0.69

- **changed** — Updated @ai-sdk/gateway to version 4.0.55

###### Patch Changes

-   Updated dependencies [1f7a464]
    -   @ai-sdk/gateway@4.0.55

### 7.0.68

- Date: 2026-08-18
- Version: 7.0.68
- Original notes: https://github.com/vercel/ai/releases/tag/ai%407.0.68
- Permalink: https://whatsnew.fyi/product/ai-sdk/releases/7.0.68

- **changed** — Updated @ai-sdk/gateway to version 4.0.54

###### Patch Changes

-   Updated dependencies [257632b]
    -   @ai-sdk/gateway@4.0.54

### 7.0.67

- Date: 2026-08-18
- Version: 7.0.67
- Original notes: https://github.com/vercel/ai/releases/tag/ai%407.0.67
- Permalink: https://whatsnew.fyi/product/ai-sdk/releases/7.0.67

- **changed** — Updated @ai-sdk/gateway to version 4.0.53

###### Patch Changes

-   Updated dependencies [a0b1ffc]
    -   @ai-sdk/gateway@4.0.53

### 7.0.66

- Date: 2026-08-14
- Version: 7.0.66
- Original notes: https://github.com/vercel/ai/releases/tag/ai%407.0.66
- Permalink: https://whatsnew.fyi/product/ai-sdk/releases/7.0.66

- **fixed** — Keep chat status submitted until response content begins streaming
- **fixed** — Fix declaration emit for exported values that infer an Output type

###### Patch Changes

-   0782259: Keep chat status submitted until response content begins streaming.
-   2fd1214: Fix declaration emit for exported values that infer an `Output` type.

### 7.0.65

- Date: 2026-08-13
- Version: 7.0.65
- Original notes: https://github.com/vercel/ai/releases/tag/ai%407.0.65
- Permalink: https://whatsnew.fyi/product/ai-sdk/releases/7.0.65

- **fixed** — Avoid repeatedly cloning accumulated text in readUIMessageStream while preserving independent snapshots for mutable nested values
- **fixed** — Preserve root-level JSON Schema definitions when wrapping array output schemas
- **fixed** — Propagate errors thrown by the Chat onFinish callback to the initiating request

###### Patch Changes

-   dc8caae: Avoid repeatedly cloning accumulated text in `readUIMessageStream` while
    preserving independent snapshots for mutable nested values.
-   72ec74f: Preserve root-level JSON Schema definitions when wrapping array output schemas.
-   c5b0515: Propagate errors thrown by the Chat `onFinish` callback to the initiating request.
-   Updated dependencies [16650e9]
    -   @ai-sdk/gateway@4.0.52

### 7.0.64

- Date: 2026-08-12
- Version: 7.0.64
- Original notes: https://github.com/vercel/ai/releases/tag/ai%407.0.64
- Permalink: https://whatsnew.fyi/product/ai-sdk/releases/7.0.64

- **changed** — @ai-sdk/gateway dependency updated to version 4.0.51

###### Patch Changes

-   Updated dependencies [ea75787]
-   Updated dependencies [b20de9e]
    -   @ai-sdk/gateway@4.0.51

### 7.0.63

- Date: 2026-08-12
- Version: 7.0.63
- Original notes: https://github.com/vercel/ai/releases/tag/ai%407.0.63
- Permalink: https://whatsnew.fyi/product/ai-sdk/releases/7.0.63

- **fixed** — Preserve reasoning block IDs from UI message streams on reasoning UI parts
- **changed** — Allow providers without reranking model support to be assigned to Provider

###### Patch Changes

-   d0a5807: Preserve reasoning block IDs from UI message streams on reasoning UI parts.
-   dcf33e8: Allow providers without reranking model support to be assigned to `Provider`.
-   Updated dependencies [a4d386d]
    -   @ai-sdk/gateway@4.0.50

### 7.0.62

- Date: 2026-08-11
- Version: 7.0.62
- Original notes: https://github.com/vercel/ai/releases/tag/ai%407.0.62
- Permalink: https://whatsnew.fyi/product/ai-sdk/releases/7.0.62

- **added** — Add typed custom bodies to Completion APIs

###### Patch Changes

-   e0bcf52: feat(ui): add typed custom bodies to Completion APIs

### 7.0.59

- Date: 2026-08-10
- Version: 7.0.59
- Original notes: https://github.com/vercel/ai/releases/tag/ai%407.0.59
- Permalink: https://whatsnew.fyi/product/ai-sdk/releases/7.0.59

###### Patch Changes

-   Updated dependencies [401a4ba]
-   Updated dependencies [7af9646]
    -   @ai-sdk/provider-utils@5.0.26
    -   @ai-sdk/gateway@4.0.47

### 7.0.58

- Date: 2026-08-07
- Version: 7.0.58
- Original notes: https://github.com/vercel/ai/releases/tag/ai%407.0.58
- Permalink: https://whatsnew.fyi/product/ai-sdk/releases/7.0.58

- **fixed** — Respect ToolLoopAgent timeouts configured in agent settings.
- **added** — Allow `aspectRatio: 'adaptive'` on `generateVideo` for video models that derive output ratio from input.
- **changed** — Reduce bundle size by making internal Zod v4 imports tree-shakeable.

###### Patch Changes

-   72ad23f: Respect ToolLoopAgent timeouts configured in agent settings.

-   ad6a650: feat(video): allow `aspectRatio: 'adaptive'` on `generateVideo`

    Some video models derive the output ratio from the input and reject explicit
    `{width}:{height}` values — BytePlus Seedance 2.5 does this for first-frame,
    first-and-last-frame, editing, and extension tasks. `aspectRatio` on
    `VideoModelV3CallOptions`, `VideoModelV4CallOptions`, and
    `experimental_generateVideo` is now `` `${number}:${number}` | 'adaptive' ``, so
    those calls no longer need a type assertion. Support is provider-specific.

-   81cd026: Reduce bundle size by making internal Zod v4 imports tree-shakeable.

-   Updated dependencies [c477556]

-   Updated dependencies [ad6a650]

-   Updated dependencies [81cd026]
    -   @ai-sdk/gateway@4.0.46
    -   @ai-sdk/provider@4.0.7
    -   @ai-sdk/provider-utils@5.0.25

### 7.0.57

- Date: 2026-08-07
- Version: 7.0.57
- Original notes: https://github.com/vercel/ai/releases/tag/ai%407.0.57
- Permalink: https://whatsnew.fyi/product/ai-sdk/releases/7.0.57

- **changed** — Updated @ai-sdk/provider-utils to 5.0.24
- **changed** — Updated @ai-sdk/gateway to 4.0.45

###### Patch Changes

-   Updated dependencies [1937bef]
    -   @ai-sdk/provider-utils@5.0.24
    -   @ai-sdk/gateway@4.0.45

### 7.0.56

- Date: 2026-08-07
- Version: 7.0.56
- Original notes: https://github.com/vercel/ai/releases/tag/ai%407.0.56
- Permalink: https://whatsnew.fyi/product/ai-sdk/releases/7.0.56

- **added** — Expose provider metadata on language-model-call end callbacks and telemetry spans
- **fixed** — Make retried doStart calls idempotent by minting one idempotency token per logical start and forwarding it as an idempotency-key header to prevent duplicate billable generations on retry after lost response
- **fixed** — Stop pending and active resumed chat streams after cancellation and prevent overlapping resumptions from applying stale updates

###### Patch Changes

-   25c9120: Expose provider metadata on language-model-call end callbacks and telemetry spans.

-   89080c8: fix (ai/gateway): make retried `doStart` calls idempotent

    `generateVideo` retries `doStart`, which creates a billable generation, so a
    retry after a lost response could start a second one. It now mints one
    idempotency token per logical start — outside the retry closure — and forwards it
    as an `idempotency-key` header, so a provider that deduplicates (the Vercel AI
    Gateway does) sees the same key on every attempt. `GatewayVideoModel` simply
    forwards the caller's headers rather than inferring retry identity from an
    options object, which would collide across unrelated calls.

-   79d6195: Stop pending and active resumed chat streams after cancellation, and prevent
    overlapping resumptions from applying stale updates.

-   Updated dependencies [89080c8]

-   Updated dependencies [89080c8]
    -   @ai-sdk/gateway@4.0.44

### 7.0.55

- Date: 2026-08-06
- Version: 7.0.55
- Original notes: https://github.com/vercel/ai/releases/tag/ai%407.0.55
- Permalink: https://whatsnew.fyi/product/ai-sdk/releases/7.0.55

- **added** — Add batch APIs

###### Patch Changes

-   3469d0c: feat: add batch APIs
-   Updated dependencies [3469d0c]
    -   @ai-sdk/provider@4.0.6
    -   @ai-sdk/gateway@4.0.43
    -   @ai-sdk/provider-utils@5.0.23

### 7.0.54

- Date: 2026-08-06
- Version: 7.0.54
- Original notes: https://github.com/vercel/ai/releases/tag/ai%407.0.54
- Permalink: https://whatsnew.fyi/product/ai-sdk/releases/7.0.54

- **added** — Allow `ToolLoopAgent` `prepareCall` callbacks to read and override the top-level `reasoning` option
- **added** — Add `defaultInstructionsMiddleware` for applying default language model instructions while preserving call-level overrides
- **fixed** — Preserve preceding assistant messages when regenerating a response

###### Patch Changes

-   a6b17a2: Allow `ToolLoopAgent` `prepareCall` callbacks to read and override the top-level `reasoning` option.
-   5615eb7: Add `defaultInstructionsMiddleware` for applying default language model instructions while preserving call-level overrides.
-   36a3ff6: Preserve preceding assistant messages when regenerating a response.

### 7.0.52

- Date: 2026-08-04
- Version: 7.0.52
- Original notes: https://github.com/vercel/ai/releases/tag/ai%407.0.52
- Permalink: https://whatsnew.fyi/product/ai-sdk/releases/7.0.52

- **fixed** — Skip re-validating tool input for terminal output-available UI message parts

###### Patch Changes

-   3836a85: Skip re-validating tool input for terminal output-available UI message parts.
-   Updated dependencies [1bec07d]
-   Updated dependencies [53c326e]
-   Updated dependencies [d765f82]
    -   @ai-sdk/provider-utils@5.0.21
    -   @ai-sdk/gateway@4.0.41

### 7.0.51

- Date: 2026-08-04
- Version: 7.0.51
- Original notes: https://github.com/vercel/ai/releases/tag/ai%407.0.51
- Permalink: https://whatsnew.fyi/product/ai-sdk/releases/7.0.51

###### Patch Changes

-   Updated dependencies [160ccdb]
    -   @ai-sdk/provider-utils@5.0.20
    -   @ai-sdk/gateway@4.0.40

### 7.0.50

- Date: 2026-08-03
- Version: 7.0.50
- Original notes: https://github.com/vercel/ai/releases/tag/ai%407.0.50
- Permalink: https://whatsnew.fyi/product/ai-sdk/releases/7.0.50

- **added** — Add asynchronous start/status flow to the experimental video model interface (VideoModelV4) with doStart, doStatus, and handleWebhookOption methods
- **added** — Add poll and webhook options to experimental_generateVideo for asynchronous completion orchestration
- **added** — Add custom delay implementation support for polling configuration to enable durable workflow compatibility
- **changed** — Simplify tool caller configuration in code-mode

###### Patch Changes

-   79e133c: async APIs for generateVideo (poll, webhook)

    Adds an asynchronous start/status flow to the experimental video model
    interface (`VideoModelV4`): models may now implement `doStart`, `doStatus`,
    and `handleWebhookOption` instead of (or in addition to) `doGenerate`, and
    `experimental_generateVideo` accepts `poll` and `webhook` options to
    orchestrate completion via polling or webhooks. Polling configuration can use
    a custom delay implementation for durable workflow compatibility.

-   da64b51: feat(code-mode): simplify tool caller configuration

-   Updated dependencies [79e133c]
    -   @ai-sdk/provider@4.0.5
    -   @ai-sdk/gateway@4.0.39
    -   @ai-sdk/provider-utils@5.0.19

### 7.0.49

- Date: 2026-08-03
- Version: 7.0.49
- Original notes: https://github.com/vercel/ai/releases/tag/ai%407.0.49
- Permalink: https://whatsnew.fyi/product/ai-sdk/releases/7.0.49

- **changed** — Updated @ai-sdk/gateway to version 4.0.38

###### Patch Changes

-   Updated dependencies [fb6d2f8]
    -   @ai-sdk/gateway@4.0.38

### 7.0.48

- Date: 2026-08-01
- Version: 7.0.48
- Original notes: https://github.com/vercel/ai/releases/tag/ai%407.0.48
- Permalink: https://whatsnew.fyi/product/ai-sdk/releases/7.0.48

- **changed** — @ai-sdk/gateway updated to version 4.0.37

###### Patch Changes

-   Updated dependencies [bdd5e28]
    -   @ai-sdk/gateway@4.0.37

### 7.0.47

- Date: 2026-07-31
- Version: 7.0.47
- Original notes: https://github.com/vercel/ai/releases/tag/ai%407.0.47
- Permalink: https://whatsnew.fyi/product/ai-sdk/releases/7.0.47

- **changed** — Updated @ai-sdk/provider-utils to version 5.0.18
- **changed** — Updated @ai-sdk/gateway to version 4.0.36

###### Patch Changes

-   Updated dependencies [5fc7da5]
-   Updated dependencies [93b2acd]
    -   @ai-sdk/provider-utils@5.0.18
    -   @ai-sdk/gateway@4.0.36

### 7.0.46

- Date: 2026-07-31
- Version: 7.0.46
- Original notes: https://github.com/vercel/ai/releases/tag/ai%407.0.46
- Permalink: https://whatsnew.fyi/product/ai-sdk/releases/7.0.46

- **changed** — Updated @ai-sdk/gateway to version 4.0.35

###### Patch Changes

-   Updated dependencies [4f2e064]
    -   @ai-sdk/gateway@4.0.35

### 7.0.45

- Date: 2026-07-31
- Version: 7.0.45
- Original notes: https://github.com/vercel/ai/releases/tag/ai%407.0.45
- Permalink: https://whatsnew.fyi/product/ai-sdk/releases/7.0.45

- **added** — Support experimental_toolCallers in streamText
- **added** — Support experimental tool callers in ToolLoopAgent
- **fixed** — Warn when generateText receives streaming-only firstChunkMs or chunkMs timeout settings

###### Patch Changes

-   d6ce0ee: feat(ai): support experimental_toolCallers in streamText
-   fa95504: feat(ai): support experimental tool callers in ToolLoopAgent
-   349afe7: Warn when `generateText` receives streaming-only `firstChunkMs` or `chunkMs` timeout settings.
-   Updated dependencies [fa95504]
    -   @ai-sdk/provider-utils@5.0.17
    -   @ai-sdk/gateway@4.0.34

### 7.0.44

- Date: 2026-07-31
- Version: 7.0.44
- Original notes: https://github.com/vercel/ai/releases/tag/ai%407.0.44
- Permalink: https://whatsnew.fyi/product/ai-sdk/releases/7.0.44

- **fixed** — Fix telemetry attribution for language model calls that resolve to a different response model

###### Patch Changes

-   015acb4: fix telemetry attribution for language model calls that resolve to a different response model