Sign in
← AI PulseSep 8, 2026

Wire · news · Multi-source brief

Vercel AI SDK Updates Batch Model Support, Tool Choices, and Image Result Retries

Vercel has released updates to its AI SDK, version ai@7.0.94, introducing per-request model support in batch operations, enforcing tool choices in streamed text, and enhancing retry logic for image results.

By Illumora Editorial

Source · Sep 8, 2026, 7:18 PM · On Illumora · Sep 8, 2026, 7:34 PM

Media from the primary source — shown here so you can stay on Illumora.

Synthesized from multiple allowlisted primaries on the same event. Lanes →

Brief drafted by Illumora’s editorial model from the linked primary source. Ops desk reviews flagged pieces. How we write →

Read the source →Vercel AI SDK Changelog — Release @ai-sdk/harness-cursor@1.0.16 · vercel/ai
Save

Vercel has updated its AI SDK to version ai@7.0.94, as detailed in recent changelog entries for @ai-sdk/harness-cursor@1.0.16, @ai-sdk/harness-cline@1.0.30, @ai-sdk/harness-fx@1.0.16, and @ai-sdk/harness-pi@1.0.105. These updates include new features and fixes aimed at improving model interaction and reliability within the SDK.

Key Points

  • The AI SDK now supports per-request models when performing batch operations, identified by change a4ba394.
  • Tool choices are now enforced in streamText functionality, a fix noted by change 36b3364.
  • The SDK retries unclassified empty image results and preserves retry-attempt accounting, according to change 45099da.
  • Provider-independent result retryability classification has been added.
  • Prompt blocks for Google and Google Vertex are now marked as terminal.
  • Dependencies for @ai-sdk/provider, @ai-sdk/gateway, @ai-sdk/provider-utils, @ai-sdk/alibaba, and @ai-sdk/amazon-bedrock have been updated.

Context

According to Vercel's changelog, these updates are part of routine patch changes to the AI SDK. The changes are managed through a Changesets release GitHub action, which automates the publishing of packages to npm upon merging. This process ensures continuous integration and delivery of updates to the SDK.

Why It Matters

These updates enhance the flexibility and robustness of the Vercel AI SDK. Developers can now specify different models for individual requests within a batch, which could optimize resource use and performance for varied tasks. Improved retry mechanisms for image results and enforced tool choices contribute to more reliable and predictable AI application behavior.

What To Do

  • Review the Vercel AI SDK documentation for details on implementing per-request models in batch operations.
  • Test the streamText function with tool choices to observe the enforced behavior.
  • Note the updated retry logic for image results, especially for applications handling image-based AI outputs.
  • Check for updated dependency versions for @ai-sdk/provider@4.0.11, @ai-sdk/gateway@4.0.76, @ai-sdk/provider-utils@5.0.37, @ai-sdk/alibaba@2.0.42, and @ai-sdk/amazon-bedrock@5.0.77 in your projects.

Keep Exploring

/studio?pack=foundation /techniques/output-schema /techniques/ptcf