Sign in
← AI PulseSep 16, 2026

Wire · news · Multi-source brief

Vercel AI SDK Adds TypeSafe Provider for experimental_evaluate and Mid-Conversation Tool Discovery

The Vercel AI SDK has introduced a TypeSafe provider for experimental_evaluate, enabling native Choice, Score, and Boolean questions, alongside mid-conversation tool discovery and updates for code-mode.

By Illumora Editorial

Source · Sep 16, 2026, 6:26 PM · On Illumora · Sep 16, 2026, 6:38 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/hume@3.0.42 · vercel/ai
Save

The Vercel AI SDK has released updates including a new TypeSafe provider for experimental_evaluate and enhancements to its ai package. These changes, noted in the @ai-sdk/hume@3.0.42 and @ai-sdk/klingai@4.0.44 releases, aim to expand evaluation capabilities and improve tool integration within conversations.

Key Points

  • The TypeSafe provider for experimental_evaluate supports native Choice, Score, and Boolean questions in a single request.
  • This provider also includes structured JSON rubrics, probability distributions, confidence metadata, and workflow serialization.
  • The ai package now features mid-conversation tool discovery and updates for code-mode.
  • The rawInput in output-error UI message parts has been deprecated.
  • Default Node.js downloads are protected by DNS validation and connection pinning, even when frameworks or instrumentation wrap global fetch.

Context

According to the Vercel AI SDK Changelog, these updates are part of broader releases across various SDK components. The @ai-sdk/typesafe-ai@3.0.0 release specifically introduces the TypeSafe provider, while the ai@7.0.103 update brings the tool discovery and deprecation changes.

Why It Matters

These updates offer builders more robust tools for evaluating AI model outputs and integrating dynamic tool use within conversational interfaces. The TypeSafe provider can streamline the process of assessing model responses with structured, native question types, potentially improving evaluation consistency and efficiency.

What To Do

  • Review the Vercel AI SDK documentation for details on the TypeSafe provider and experimental_evaluate.
  • Investigate the new mid-conversation tool discovery features for potential integration into existing code-mode applications.
  • Note the deprecation of rawInput in output-error UI message parts and plan for any necessary adjustments.
  • Examine how the enhanced DNS validation and connection pinning might affect custom fetch implementations.