Sign in
← AI PulseAug 4, 2026

Wire · news · Multi-source brief

Vercel AI SDK Updates Baseten Integration for Embeddings

The Vercel AI SDK has updated its integration with Baseten, making the native performance client for embeddings opt-in and removing it as a default dependency.

By Illumora Editorial

Source · Aug 4, 2026, 5:16 AM · On Illumora · Aug 4, 2026, 5:28 AM

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/vercel@3.0.22 · vercel/ai
Save

The Vercel AI SDK has released updates across several packages, including @ai-sdk/vercel@3.0.22, @ai-sdk/vue@4.0.51, @ai-sdk/voyage@2.0.20, and @ai-sdk/workflow-harness@1.0.57. These updates, detailed in the Vercel AI SDK Changelog, primarily concern the @ai-sdk/baseten@2.1.0 package, which now makes the native performance client for embeddings an opt-in feature.

Key Points

  • The @basetenlabs/performance-client is no longer a default dependency in @ai-sdk/baseten@2.1.0.
  • This client is a NAPI addon comprising 16 platform binary packages.
  • The NAPI addon previously added ~5-16 MB to installations.
  • It could not load in edge runtimes and its platform binaries were not resolvable by bundlers.
  • The client was imported at the module top level, meaning all consumers incurred its cost even if they only used embeddings.
  • Embeddings now use plain HTTP to the deployment's OpenAI-compatible endpoint.
  • This endpoint is served by Baseten Embeddings Inference without additional settings.
  • Users can still install the PerformanceClient themselves for client-side batching and request hedging.

Context

According to the Vercel AI SDK Changelog, the @basetenlabs/performance-client was identified as a dependency that caused issues due to its nature as a NAPI addon. This addon's size and inability to function in edge runtimes or with certain bundlers led to its removal as a mandatory dependency. The change aims to reduce the default installation footprint and improve compatibility across different environments.

Why It Matters

This update impacts developers using the Vercel AI SDK for embedding functionalities, offering greater control over dependencies and potentially reducing bundle sizes and deployment complexities, especially in edge runtime environments. It shifts the responsibility for performance client installation to the developer, allowing them to decide if the benefits of client-side batching and request hedging outweigh the added dependency.

What To Do

  • Note that @basetenlabs/performance-client is no longer automatically included with @ai-sdk/baseten.
  • If client-side batching or request hedging for embeddings is desired, install @basetenlabs/performance-client manually.
  • Review the updated @ai-sdk/baseten documentation for guidance on integrating the performance client.
  • Test existing embedding implementations to ensure continued functionality with the new opt-in approach.