Skip to content

ci: Version Packages - #1699

Open
github-actions[bot] wants to merge 1 commit into
mainfrom
changeset-release/main
Open

ci: Version Packages#1699
github-actions[bot] wants to merge 1 commit into
mainfrom
changeset-release/main

Conversation

@github-actions

@github-actions github-actions Bot commented Oct 9, 2026

Copy link
Copy Markdown
Contributor

This PR was opened by the Changesets release GitHub action. When you're ready to do a release, you can merge this and the packages will be published to npm automatically. If you're not ready to do a release yet, that's fine, whenever you add more changesets to main, this PR will be updated.

Releases

@tanstack/ai@0.69.0

Minor Changes

  • #1698 1bfcee3 - Keep the order of Claude's thinking, text, and tool calls in one answer.

    • Claude can answer with thinking, a tool call, more thinking, and another tool call. The stream, the stored messages, the client wire, and the next request to Claude keep that order.
    • ModelMessage has a new optional blockOrder field. The library writes it only when the order is not the default (thinking, then text, then tool calls), so a message in the default order does not change. orderedAssistantBlocks(message) gives adapter authors the blocks in order.
    • Each Claude thinking block ends with its own REASONING_MESSAGE_END and REASONING_END.
    • Text after a second thinking block starts a new text part.
    • On the client wire, an answer in another order goes out as ordered rows. A row that exists only for the order has metadata.tanstack.continues, and the server joins it back into one message. A UIMessage that holds two model calls with a tool result between them reaches the server as assistant, tool, assistant, for every adapter.

  • #1698 1bfcee3 - Keep a conversation valid when the next request goes to a different model.

    • Each assistant message records metadata.tanstack.source ({ provider, api, model }), and responseId and the resolved model when the provider sends them. A failed or aborted answer gets stopReason: 'error' or 'aborted'. RUN_FINISHED has a new responseId field. TextAdapter has new optional provider and api fields.
    • When the next request goes to a different source, the adapter removes signatures and redacted thinking, turns readable thinking into text, and rewrites tool call IDs to the shape the target API accepts. Each call stays paired with its result.
    • Every request drops failed and aborted assistant batches and adds a No result provided result for an unanswered tool call. The saved history does not change.
    • RUN_FINISHED and RUN_ERROR metadata now merge onto the assistant messages of that model call in StreamProcessor.
    • Anthropic: the new allowEmptySignature option (default false) replays unsigned thinking from the same source, for gateways. The new provider option names a gateway.
    • Azure OpenAI reports the azure-openai-responses API.
    • Gemini, Mistral, and Bedrock Converse send same-source thinking back to the model, in block order.
    • withPersistence writes the source and the stop reason onto streaming snapshot rows.
    • Adapter authors get transformMessagesForReplay() and hashToolCallId() from @tanstack/ai/adapter-internals.

@tanstack/ai-anthropic@0.22.0

Minor Changes

  • #1698 1bfcee3 - Keep the order of Claude's thinking, text, and tool calls in one answer.

    • Claude can answer with thinking, a tool call, more thinking, and another tool call. The stream, the stored messages, the client wire, and the next request to Claude keep that order.
    • ModelMessage has a new optional blockOrder field. The library writes it only when the order is not the default (thinking, then text, then tool calls), so a message in the default order does not change. orderedAssistantBlocks(message) gives adapter authors the blocks in order.
    • Each Claude thinking block ends with its own REASONING_MESSAGE_END and REASONING_END.
    • Text after a second thinking block starts a new text part.
    • On the client wire, an answer in another order goes out as ordered rows. A row that exists only for the order has metadata.tanstack.continues, and the server joins it back into one message. A UIMessage that holds two model calls with a tool result between them reaches the server as assistant, tool, assistant, for every adapter.

  • #1698 1bfcee3 - Keep a conversation valid when the next request goes to a different model.

    • Each assistant message records metadata.tanstack.source ({ provider, api, model }), and responseId and the resolved model when the provider sends them. A failed or aborted answer gets stopReason: 'error' or 'aborted'. RUN_FINISHED has a new responseId field. TextAdapter has new optional provider and api fields.
    • When the next request goes to a different source, the adapter removes signatures and redacted thinking, turns readable thinking into text, and rewrites tool call IDs to the shape the target API accepts. Each call stays paired with its result.
    • Every request drops failed and aborted assistant batches and adds a No result provided result for an unanswered tool call. The saved history does not change.
    • RUN_FINISHED and RUN_ERROR metadata now merge onto the assistant messages of that model call in StreamProcessor.
    • Anthropic: the new allowEmptySignature option (default false) replays unsigned thinking from the same source, for gateways. The new provider option names a gateway.
    • Azure OpenAI reports the azure-openai-responses API.
    • Gemini, Mistral, and Bedrock Converse send same-source thinking back to the model, in block order.
    • withPersistence writes the source and the stop reason onto streaming snapshot rows.
    • Adapter authors get transformMessagesForReplay() and hashToolCallId() from @tanstack/ai/adapter-internals.

Patch Changes

@tanstack/ai-acp@0.3.25

Patch Changes

  • Updated dependencies [1bfcee3, 1bfcee3]:
    • @tanstack/ai@0.69.0
    • @tanstack/ai-sandbox@0.5.23

@tanstack/ai-angular@0.13.4

Patch Changes

  • Updated dependencies [1bfcee3, 1bfcee3]:
    • @tanstack/ai@0.69.0
    • @tanstack/ai-client@0.39.3

@tanstack/ai-bedrock@0.5.2

Patch Changes

  • #1698 1bfcee3 - Keep a conversation valid when the next request goes to a different model.

    • Each assistant message records metadata.tanstack.source ({ provider, api, model }), and responseId and the resolved model when the provider sends them. A failed or aborted answer gets stopReason: 'error' or 'aborted'. RUN_FINISHED has a new responseId field. TextAdapter has new optional provider and api fields.
    • When the next request goes to a different source, the adapter removes signatures and redacted thinking, turns readable thinking into text, and rewrites tool call IDs to the shape the target API accepts. Each call stays paired with its result.
    • Every request drops failed and aborted assistant batches and adds a No result provided result for an unanswered tool call. The saved history does not change.
    • RUN_FINISHED and RUN_ERROR metadata now merge onto the assistant messages of that model call in StreamProcessor.
    • Anthropic: the new allowEmptySignature option (default false) replays unsigned thinking from the same source, for gateways. The new provider option names a gateway.
    • Azure OpenAI reports the azure-openai-responses API.
    • Gemini, Mistral, and Bedrock Converse send same-source thinking back to the model, in block order.
    • withPersistence writes the source and the stop reason onto streaming snapshot rows.
    • Adapter authors get transformMessagesForReplay() and hashToolCallId() from @tanstack/ai/adapter-internals.

  • Updated dependencies [1bfcee3, 1bfcee3]:

    • @tanstack/ai@0.69.0
    • @tanstack/openai-base@0.13.2

@tanstack/ai-byteplus@0.6.5

Patch Changes

  • Updated dependencies [1bfcee3, 1bfcee3]:
    • @tanstack/ai@0.69.0
    • @tanstack/openai-base@0.13.2

@tanstack/ai-claude-code@0.8.2

Patch Changes

  • Updated dependencies [1bfcee3, 1bfcee3]:
    • @tanstack/ai@0.69.0
    • @tanstack/ai-sandbox@0.5.23

@tanstack/ai-client@0.39.3

Patch Changes

@tanstack/ai-cloudflare@0.2.8

Patch Changes

  • Updated dependencies [1bfcee3, 1bfcee3]:
    • @tanstack/ai@0.69.0
    • @tanstack/openai-base@0.13.2

@tanstack/ai-code-mode@0.5.3

Patch Changes

@tanstack/ai-code-mode-snippets@0.4.25

Patch Changes

  • Updated dependencies [1bfcee3, 1bfcee3]:
    • @tanstack/ai@0.69.0
    • @tanstack/ai-code-mode@0.5.3

@tanstack/ai-codex@0.5.20

Patch Changes

  • Updated dependencies [1bfcee3, 1bfcee3]:
    • @tanstack/ai@0.69.0
    • @tanstack/ai-sandbox@0.5.23

@tanstack/ai-cohere@0.4.17

Patch Changes

@tanstack/ai-compaction@0.1.18

Patch Changes

@tanstack/ai-devtools-core@0.5.28

Patch Changes

@tanstack/ai-durable-stream@0.1.27

Patch Changes

@tanstack/ai-elevenlabs@0.6.12

Patch Changes

@tanstack/ai-fal@0.16.4

Patch Changes

@tanstack/ai-gemini@0.37.3

Patch Changes

  • #1698 1bfcee3 - Keep a conversation valid when the next request goes to a different model.

    • Each assistant message records metadata.tanstack.source ({ provider, api, model }), and responseId and the resolved model when the provider sends them. A failed or aborted answer gets stopReason: 'error' or 'aborted'. RUN_FINISHED has a new responseId field. TextAdapter has new optional provider and api fields.
    • When the next request goes to a different source, the adapter removes signatures and redacted thinking, turns readable thinking into text, and rewrites tool call IDs to the shape the target API accepts. Each call stays paired with its result.
    • Every request drops failed and aborted assistant batches and adds a No result provided result for an unanswered tool call. The saved history does not change.
    • RUN_FINISHED and RUN_ERROR metadata now merge onto the assistant messages of that model call in StreamProcessor.
    • Anthropic: the new allowEmptySignature option (default false) replays unsigned thinking from the same source, for gateways. The new provider option names a gateway.
    • Azure OpenAI reports the azure-openai-responses API.
    • Gemini, Mistral, and Bedrock Converse send same-source thinking back to the model, in block order.
    • withPersistence writes the source and the stop reason onto streaming snapshot rows.
    • Adapter authors get transformMessagesForReplay() and hashToolCallId() from @tanstack/ai/adapter-internals.

  • Updated dependencies [1bfcee3, 1bfcee3]:

    • @tanstack/ai@0.69.0

@tanstack/ai-grok@0.21.5

Patch Changes

  • Updated dependencies [1bfcee3, 1bfcee3]:
    • @tanstack/ai@0.69.0
    • @tanstack/openai-base@0.13.2

@tanstack/ai-grok-build@0.5.20

Patch Changes

  • Updated dependencies [1bfcee3, 1bfcee3]:
    • @tanstack/ai@0.69.0
    • @tanstack/ai-acp@0.3.25
    • @tanstack/ai-sandbox@0.5.23

@tanstack/ai-groq@0.8.8

Patch Changes

  • Updated dependencies [1bfcee3, 1bfcee3]:
    • @tanstack/ai@0.69.0
    • @tanstack/openai-base@0.13.2

@tanstack/ai-llmgateway@0.2.8

Patch Changes

  • Updated dependencies [1bfcee3, 1bfcee3]:
    • @tanstack/ai@0.69.0
    • @tanstack/openai-base@0.13.2

@tanstack/ai-lovable@0.5.5

Patch Changes

  • Updated dependencies [1bfcee3, 1bfcee3]:
    • @tanstack/ai@0.69.0
    • @tanstack/openai-base@0.13.2

@tanstack/ai-mcp@0.8.4

Patch Changes

@tanstack/ai-memory@0.2.14

Patch Changes

@tanstack/ai-mistral@0.6.19

Patch Changes

  • #1698 1bfcee3 - Keep a conversation valid when the next request goes to a different model.

    • Each assistant message records metadata.tanstack.source ({ provider, api, model }), and responseId and the resolved model when the provider sends them. A failed or aborted answer gets stopReason: 'error' or 'aborted'. RUN_FINISHED has a new responseId field. TextAdapter has new optional provider and api fields.
    • When the next request goes to a different source, the adapter removes signatures and redacted thinking, turns readable thinking into text, and rewrites tool call IDs to the shape the target API accepts. Each call stays paired with its result.
    • Every request drops failed and aborted assistant batches and adds a No result provided result for an unanswered tool call. The saved history does not change.
    • RUN_FINISHED and RUN_ERROR metadata now merge onto the assistant messages of that model call in StreamProcessor.
    • Anthropic: the new allowEmptySignature option (default false) replays unsigned thinking from the same source, for gateways. The new provider option names a gateway.
    • Azure OpenAI reports the azure-openai-responses API.
    • Gemini, Mistral, and Bedrock Converse send same-source thinking back to the model, in block order.
    • withPersistence writes the source and the stop reason onto streaming snapshot rows.
    • Adapter authors get transformMessagesForReplay() and hashToolCallId() from @tanstack/ai/adapter-internals.

  • Updated dependencies [1bfcee3, 1bfcee3]:

    • @tanstack/ai@0.69.0

@tanstack/ai-octane@0.7.4

Patch Changes

  • Updated dependencies [1bfcee3, 1bfcee3]:
    • @tanstack/ai@0.69.0
    • @tanstack/ai-client@0.39.3

@tanstack/ai-ollama@0.11.17

Patch Changes

  • #1698 1bfcee3 - Keep a conversation valid when the next request goes to a different model.

    • Each assistant message records metadata.tanstack.source ({ provider, api, model }), and responseId and the resolved model when the provider sends them. A failed or aborted answer gets stopReason: 'error' or 'aborted'. RUN_FINISHED has a new responseId field. TextAdapter has new optional provider and api fields.
    • When the next request goes to a different source, the adapter removes signatures and redacted thinking, turns readable thinking into text, and rewrites tool call IDs to the shape the target API accepts. Each call stays paired with its result.
    • Every request drops failed and aborted assistant batches and adds a No result provided result for an unanswered tool call. The saved history does not change.
    • RUN_FINISHED and RUN_ERROR metadata now merge onto the assistant messages of that model call in StreamProcessor.
    • Anthropic: the new allowEmptySignature option (default false) replays unsigned thinking from the same source, for gateways. The new provider option names a gateway.
    • Azure OpenAI reports the azure-openai-responses API.
    • Gemini, Mistral, and Bedrock Converse send same-source thinking back to the model, in block order.
    • withPersistence writes the source and the stop reason onto streaming snapshot rows.
    • Adapter authors get transformMessagesForReplay() and hashToolCallId() from @tanstack/ai/adapter-internals.

  • Updated dependencies [1bfcee3, 1bfcee3]:

    • @tanstack/ai@0.69.0

@tanstack/ai-ollaya@0.2.5

Patch Changes

@tanstack/ai-openai@0.29.2

Patch Changes

  • #1698 1bfcee3 - Keep a conversation valid when the next request goes to a different model.

    • Each assistant message records metadata.tanstack.source ({ provider, api, model }), and responseId and the resolved model when the provider sends them. A failed or aborted answer gets stopReason: 'error' or 'aborted'. RUN_FINISHED has a new responseId field. TextAdapter has new optional provider and api fields.
    • When the next request goes to a different source, the adapter removes signatures and redacted thinking, turns readable thinking into text, and rewrites tool call IDs to the shape the target API accepts. Each call stays paired with its result.
    • Every request drops failed and aborted assistant batches and adds a No result provided result for an unanswered tool call. The saved history does not change.
    • RUN_FINISHED and RUN_ERROR metadata now merge onto the assistant messages of that model call in StreamProcessor.
    • Anthropic: the new allowEmptySignature option (default false) replays unsigned thinking from the same source, for gateways. The new provider option names a gateway.
    • Azure OpenAI reports the azure-openai-responses API.
    • Gemini, Mistral, and Bedrock Converse send same-source thinking back to the model, in block order.
    • withPersistence writes the source and the stop reason onto streaming snapshot rows.
    • Adapter authors get transformMessagesForReplay() and hashToolCallId() from @tanstack/ai/adapter-internals.

  • Updated dependencies [1bfcee3, 1bfcee3]:

    • @tanstack/ai@0.69.0
    • @tanstack/openai-base@0.13.2

@tanstack/ai-opencode@0.4.20

Patch Changes

  • Updated dependencies [1bfcee3, 1bfcee3]:
    • @tanstack/ai@0.69.0
    • @tanstack/ai-sandbox@0.5.23

@tanstack/ai-openrouter@0.21.6

Patch Changes

  • #1698 1bfcee3 - Keep a conversation valid when the next request goes to a different model.

    • Each assistant message records metadata.tanstack.source ({ provider, api, model }), and responseId and the resolved model when the provider sends them. A failed or aborted answer gets stopReason: 'error' or 'aborted'. RUN_FINISHED has a new responseId field. TextAdapter has new optional provider and api fields.
    • When the next request goes to a different source, the adapter removes signatures and redacted thinking, turns readable thinking into text, and rewrites tool call IDs to the shape the target API accepts. Each call stays paired with its result.
    • Every request drops failed and aborted assistant batches and adds a No result provided result for an unanswered tool call. The saved history does not change.
    • RUN_FINISHED and RUN_ERROR metadata now merge onto the assistant messages of that model call in StreamProcessor.
    • Anthropic: the new allowEmptySignature option (default false) replays unsigned thinking from the same source, for gateways. The new provider option names a gateway.
    • Azure OpenAI reports the azure-openai-responses API.
    • Gemini, Mistral, and Bedrock Converse send same-source thinking back to the model, in block order.
    • withPersistence writes the source and the stop reason onto streaming snapshot rows.
    • Adapter authors get transformMessagesForReplay() and hashToolCallId() from @tanstack/ai/adapter-internals.

  • Updated dependencies [1bfcee3, 1bfcee3]:

    • @tanstack/ai@0.69.0

@tanstack/ai-perplexity@0.2.25

Patch Changes

@tanstack/ai-persistence@0.8.4

Patch Changes

  • #1698 1bfcee3 - Keep a conversation valid when the next request goes to a different model.

    • Each assistant message records metadata.tanstack.source ({ provider, api, model }), and responseId and the resolved model when the provider sends them. A failed or aborted answer gets stopReason: 'error' or 'aborted'. RUN_FINISHED has a new responseId field. TextAdapter has new optional provider and api fields.
    • When the next request goes to a different source, the adapter removes signatures and redacted thinking, turns readable thinking into text, and rewrites tool call IDs to the shape the target API accepts. Each call stays paired with its result.
    • Every request drops failed and aborted assistant batches and adds a No result provided result for an unanswered tool call. The saved history does not change.
    • RUN_FINISHED and RUN_ERROR metadata now merge onto the assistant messages of that model call in StreamProcessor.
    • Anthropic: the new allowEmptySignature option (default false) replays unsigned thinking from the same source, for gateways. The new provider option names a gateway.
    • Azure OpenAI reports the azure-openai-responses API.
    • Gemini, Mistral, and Bedrock Converse send same-source thinking back to the model, in block order.
    • withPersistence writes the source and the stop reason onto streaming snapshot rows.
    • Adapter authors get transformMessagesForReplay() and hashToolCallId() from @tanstack/ai/adapter-internals.

  • Updated dependencies [1bfcee3, 1bfcee3]:

    • @tanstack/ai@0.69.0

@tanstack/ai-preact@0.20.4

Patch Changes

  • Updated dependencies [1bfcee3, 1bfcee3]:
    • @tanstack/ai@0.69.0
    • @tanstack/ai-client@0.39.3

@tanstack/ai-react@0.30.4

Patch Changes

  • Updated dependencies [1bfcee3, 1bfcee3]:
    • @tanstack/ai@0.69.0
    • @tanstack/ai-client@0.39.3

@tanstack/ai-reactor@0.2.15

Patch Changes

@tanstack/ai-remix@0.6.4

Patch Changes

  • Updated dependencies [1bfcee3, 1bfcee3]:
    • @tanstack/ai@0.69.0
    • @tanstack/ai-client@0.39.3

@tanstack/ai-sandbox@0.5.23

Patch Changes

  • Updated dependencies [1bfcee3, 1bfcee3]:
    • @tanstack/ai@0.69.0
    • @tanstack/ai-persistence@0.8.4
    • @tanstack/ai-skills@0.1.19

@tanstack/ai-sandbox-cloudflare@0.4.15

Patch Changes

  • Updated dependencies [1bfcee3, 1bfcee3]:
    • @tanstack/ai@0.69.0
    • @tanstack/ai-sandbox@0.5.23
    • @tanstack/ai-sandbox-local-process@0.2.6

@tanstack/ai-skills@0.1.19

Patch Changes

@tanstack/ai-solid@0.26.4

Patch Changes

  • Updated dependencies [1bfcee3, 1bfcee3]:
    • @tanstack/ai@0.69.0
    • @tanstack/ai-client@0.39.3

@tanstack/ai-svelte@0.25.4

Patch Changes

  • Updated dependencies [1bfcee3, 1bfcee3]:
    • @tanstack/ai@0.69.0
    • @tanstack/ai-client@0.39.3

@tanstack/ai-typesafe@0.1.12

Patch Changes

@tanstack/ai-vercel-gateway@0.3.10

Patch Changes

  • Updated dependencies [1bfcee3, 1bfcee3]:
    • @tanstack/ai@0.69.0
    • @tanstack/openai-base@0.13.2

@tanstack/ai-vertex@0.2.29

Patch Changes

  • Updated dependencies [1bfcee3, 1bfcee3]:
    • @tanstack/ai@0.69.0
    • @tanstack/ai-gemini@0.37.3

@tanstack/ai-vue@0.26.4

Patch Changes

  • Updated dependencies [1bfcee3, 1bfcee3]:
    • @tanstack/ai@0.69.0
    • @tanstack/ai-client@0.39.3

@tanstack/ai-worldlabs@0.1.8

Patch Changes

@tanstack/openai-base@0.13.2

Patch Changes

  • #1698 1bfcee3 - Keep a conversation valid when the next request goes to a different model.

    • Each assistant message records metadata.tanstack.source ({ provider, api, model }), and responseId and the resolved model when the provider sends them. A failed or aborted answer gets stopReason: 'error' or 'aborted'. RUN_FINISHED has a new responseId field. TextAdapter has new optional provider and api fields.
    • When the next request goes to a different source, the adapter removes signatures and redacted thinking, turns readable thinking into text, and rewrites tool call IDs to the shape the target API accepts. Each call stays paired with its result.
    • Every request drops failed and aborted assistant batches and adds a No result provided result for an unanswered tool call. The saved history does not change.
    • RUN_FINISHED and RUN_ERROR metadata now merge onto the assistant messages of that model call in StreamProcessor.
    • Anthropic: the new allowEmptySignature option (default false) replays unsigned thinking from the same source, for gateways. The new provider option names a gateway.
    • Azure OpenAI reports the azure-openai-responses API.
    • Gemini, Mistral, and Bedrock Converse send same-source thinking back to the model, in block order.
    • withPersistence writes the source and the stop reason onto streaming snapshot rows.
    • Adapter authors get transformMessagesForReplay() and hashToolCallId() from @tanstack/ai/adapter-internals.

  • Updated dependencies [1bfcee3, 1bfcee3]:

    • @tanstack/ai@0.69.0

ag-ui@0.0.38

Patch Changes

  • Updated dependencies []:
    • @tanstack/ai-client@0.39.3
    • @tanstack/ai-react@0.30.4

@coderabbitai

coderabbitai Bot commented Oct 9, 2026

Copy link
Copy Markdown
Contributor

Important

Review skipped

Bot user detected.

To trigger a single review, invoke the @coderabbitai review command.

鈿欙笍 Run configuration
  • Configuration used: Repository: TanStack/ai/.coderabbit.yaml
  • Review profile: CHILL
  • Plan: Advanced
  • Run ID: 00e37fe1-784f-4f14-b126-cf2aea481278

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

  • 馃攳 Trigger review
  • Autofix 路 Keep fixing CodeRabbit findings and required CI, and resolving merge conflicts

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.

鉂わ笍 Share

Comment @coderabbitai help to get the list of available commands.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

No reviews

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

0 participants