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 an opt-in feature in version @ai-sdk/baseten@2.1.0.

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/svelte@5.0.51 · vercel/ai
Save

The Vercel AI SDK has released an update to its @ai-sdk/baseten package, version 2.1.0, which alters how the native performance client for embeddings is handled. This change makes the client an opt-in component, addressing previous issues with its integration.

Key Points

  • The @ai-sdk/baseten@2.1.0 package makes the native performance client opt-in for embeddings.
  • @basetenlabs/performance-client is no longer a direct dependency of the @ai-sdk/baseten package.
  • The native client is a NAPI addon, comprising 16 platform binary packages and requiring 5-16 MB of installation space.
  • This addon previously could not load in edge runtimes, and bundlers struggled to resolve its platform binaries.
  • The native client was imported at the module top level, meaning all consumers incurred its overhead even if they only used embeddings.
  • Embeddings now utilize plain HTTP to an OpenAI-compatible endpoint, which Baseten Embeddings Inference serves without additional configuration.
  • Users can maintain client-side batching and request hedging by manually installing @basetenlabs/performance-client and passing its constructor.

Context

According to the Vercel AI SDK changelog, the previous implementation of the native performance client presented challenges. Its nature as a NAPI addon led to compatibility issues with edge runtimes and difficulties for bundlers. Furthermore, its top-level import meant that all users of the @ai-sdk/baseten package bore the cost of its dependency, regardless of whether they used its embedding features.

Why It Matters

This update from Vercel AI SDK simplifies the dependency structure for developers using @ai-sdk/baseten, potentially reducing bundle sizes and improving compatibility with various deployment environments, especially edge runtimes. It allows developers to choose whether to include the native performance client based on their specific needs for embedding features.

What To Do

  • Developers using @ai-sdk/baseten should note the change in dependency for the native performance client.
  • If client-side batching and request hedging are desired for embeddings, install @basetenlabs/performance-client separately.
  • Review the updated @ai-sdk/baseten documentation for guidance on integrating the performance client.
  • Test applications in edge runtimes to confirm improved compatibility following this update.