Sign in
← AI PulseSep 16, 2026

Wire · news · Multi-source brief

Vercel AI SDK Adds TypeSafe Provider for Evaluation, Enhances Tool Discovery

The Vercel AI SDK has introduced a TypeSafe provider for experimental_evaluate, enabling native Choice, Score, and Boolean questions, structured JSON rubrics, and workflow serialization.

By Illumora Editorial

Source · Sep 16, 2026, 6:26 PM · On Illumora · Sep 16, 2026, 6:39 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-opencode@1.0.115 · vercel/ai
Save

The Vercel AI SDK has released updates including a new TypeSafe provider for experimental_evaluate. This provider supports native Choice, Score, and Boolean questions within a single request, along with structured JSON rubrics, probability distributions, confidence metadata, and workflow serialization. These changes are part of the @ai-sdk/typesafe-ai@3.0.0 major release, as noted in the Vercel AI SDK Changelog.

Key Points

  • The TypeSafe provider for experimental_evaluate is now available in the Vercel AI SDK.
  • It supports native Choice, Score, and Boolean questions in one request.
  • The provider includes features for structured JSON rubrics, probability distributions, and confidence metadata.
  • Workflow serialization is also supported by the new TypeSafe provider.
  • Mid-conversation tool discovery and updates for code-mode have been added to the ai package.
  • The rawInput field in output-error UI message parts has been deprecated.
  • Node.js downloads are protected by DNS validation and connection pinning even when fetch is wrapped.

Context

According to the Vercel AI SDK Changelog, the @ai-sdk/typesafe-ai@3.0.0 release introduces significant evaluation capabilities. Additionally, the ai@7.0.103 package includes enhancements for tool discovery during conversations and deprecates a specific UI message part. The changelog also highlights continued security measures for Node.js downloads.

Why It Matters

These updates provide developers using the Vercel AI SDK with more robust tools for evaluating AI model outputs and integrating dynamic tool use. The structured evaluation features can streamline the assessment of model performance, while improved tool discovery can lead to more interactive and capable AI applications.

What To Do

  • Review the Vercel AI SDK documentation for details on the new TypeSafe provider and experimental_evaluate.
  • Explore the capabilities of native Choice, Score, and Boolean questions for evaluation.
  • Investigate the use of structured JSON rubrics and workflow serialization in your evaluation pipelines.
  • Note the deprecation of rawInput in output-error UI message parts and adjust code accordingly.
  • Test the new mid-conversation tool discovery features for applications using code-mode.