Sign in
← AI PulseSep 16, 2026

Wire · news · Multi-source brief

Vercel AI SDK Introduces Experimental Evaluation Model Aliases and Tool Search

The Vercel AI SDK, in its ai@7.0.104 patch release, has added experimental evaluation model aliases and registry resolution, alongside new features for tool search and direct tool calling.

By Illumora Editorial

Source · Sep 16, 2026, 9:31 PM · On Illumora · Sep 16, 2026, 9:43 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/xai@5.0.2: Version Packages (#20860) · vercel/ai
Save

Vercel has released patch ai@7.0.104 for its AI SDK, introducing experimental evaluation model aliases and registry resolution. This update also includes new functionalities for tool search, supporting both direct tool calling and native tool search capabilities.

Key Points

  • The ai@7.0.104 patch release adds experimental evaluation model aliases and registry resolution.
  • The customProvider now accepts evaluationModels.
  • Registries expose evaluationModel for resolution.
  • The experimental_evaluate function accepts string IDs when an evaluation-capable default provider is explicitly configured.
  • Evaluation will not implicitly fall back to Gateway.
  • Model-resolution errors now identify evaluationModel.
  • The SDK supports tool search with direct tool calling.
  • Native tool search tool functionality has been added.

Context

According to the Vercel AI SDK Changelog, the ai@7.0.104 patch release also includes a fix to report abnormal realtime WebSocket close diagnostics through onError. This release updates dependencies for @ai-sdk/provider-utils@5.0.43, @ai-sdk/provider@4.0.17, and @ai-sdk/gateway@4.

Why It Matters

These updates provide developers with more granular control over model evaluation and enhance the SDK's capabilities for integrating external tools, potentially streamlining development workflows and improving diagnostic insights.

What To Do

  • Review the ai@7.0.104 patch notes for detailed changes.
  • Test the experimental_evaluate function with explicitly configured evaluation-capable providers.
  • Explore the new tool search features for direct and native tool calling in your applications.
  • Note the changes to customProvider and registry exposures for evaluationModel.

Keep Exploring

/studio?pack=foundation