Major Changes

• 832f86f: fix(anthropic): remove cacheCreationInputTokens from providerMetadata

• ef992f8: Remove CommonJS exports from all packages. All packages are now ESM-only ("type": "module"). Consumers using require() must switch to ESM import syntax.

• 8359612: Start v7 pre-release

• 04e9009: chore: 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

• e02f041: feat(provider/anthropic): add support for claude-opus-4-8

• 38fc777: Add AI Gateway hint to provider READMEs

• e748b35: chore: update v3 specs to v4

• 00a0f36: feat(google, google-vertex): added gemini-embedding-2, deep-research-max-preview-04-2026 and deep-research-preview-04-2026

• 42bdfa4: fix(google-vertex): propagate project-id set in provider instance creation

• e59c955: feat(vertex): add grok models to vertex provider

• 6dd2fd7: feat(provider/google-vertex): add Google Cloud Speech-to-Text transcription model support

• 2ce3c65: feat(provider/google-vertex): add Gemini text-to-speech (speech) model support

• a3757d7: chore(provider/google): remove useless GenerativeAI affix from types and classes for consistency

• 87d1723: chore(anthropic): remove unnecessary messages affix from Anthropic symbols and files

• 9f0e36c: trigger release for all packages after provenance setup

• 1f4058f: fix(vertex): use correct import for token generator

• 82288b0: feat(provider/google): add gemini-embedding-2-preview and fix multimodal embedding support with embedMany

• e59176c: chore(google): update v3 specs to v4

• ec2e752: fix(google-vertex): support regional endpoints for vertex MaaS

• bb93832: fix(vertex-anthropic): support eu/us multi-region anthropic endpoints

• 7fc6bd6: Raise minimum supported Node.js version to 22. Supported versions: 22, 24, and 26.

• 0c4c275: trigger initial canary release

• ebb02ea: feat(google-vertex): add the anthropic tool search tools

• b70f6ec: fix(google-vertex): support for eu/us multi region gemini endpoints

• 9bd6512: feat(provider): change file part data property to be tagged with a type and remove the image part type

• 258c093: chore: ensure consistent import handling and avoid import duplicates or cycles

• 546cefe: feat(provider/google): add gemini-3.5-flash

• 55f0938: chore(provider/google): update available models

• b8396f0: trigger initial beta release

• 96d056d: fix(provider/google-vertex): avoid recreating Node GoogleAuth clients for repeated requests

  Create Google auth token generators per provider instance instead of using a module-level shared GoogleAuth cache. This avoids unnecessary GoogleAuth recreation when googleAuthOptions are omitted or when multiple provider instances use equivalent auth settings.

• 6b4d325: feat(provider/anthropic): add support for claude-fable-5 and the fallbacks API parameter

• f05a40d: fix(vertex): throw warning when strict: true for vertexAnthropic

• 90e2d8a: chore: fix unused vars not being flagged by our lint tooling

• a34b997: feat (provider/google-vertex): add support for maas models

• 780e563: feat(provider/google-vertex): allow overriding Vertex Anthropic auth token generation

• 427a596: Route gemini-embedding-2 / gemini-embedding-2-preview to the :embedContent endpoint, which is the only one those models support (:predict returns 400 FAILED_PRECONDITION)

• 0416e3e: feat (video): add first-class generateAudio call option

• b3976a2: Add workflow serialization support to all provider models.

  @ai-sdk/provider-utils: New serializeModel() 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: headers is now optional in provider config types. This is non-breaking — existing code that passes headers continues to work. Custom provider implementations that construct model configs manually can now omit headers, which is useful when models are deserialized from a workflow step boundary where auth is provided separately.

  All provider model classes now include WORKFLOW_SERIALIZE and WORKFLOW_DESERIALIZE static methods, enabling them to cross workflow step boundaries without serialization errors.

• 0d8f107: feat(provider/anthropic): add support for Opus 4.7 and relevant API enhancements