Sign in
← AI PulseAug 30, 2026

Wire · news · Multi-source brief

Vercel AI SDK Updates Anthropic and Amazon Bedrock Integrations

The Vercel AI SDK has received updates to its Anthropic and Amazon Bedrock integrations, including support for Anthropic reasoning budgets and Amazon Bedrock streaming responses.

By Illumora Editorial

Source · Aug 30, 2026, 7:58 AM · On Illumora · Aug 30, 2026, 8:08 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/togetherai@3.0.42 · vercel/ai
Save

The Vercel AI SDK has released several updates across its packages, including enhancements for Anthropic and Amazon Bedrock integrations. These changes were published as part of releases for @ai-sdk/togetherai@3.0.42 and @ai-sdk/tui@1.0.86, as well as earlier releases like @ai-sdk/togetherai@3.0.41 and @ai-sdk/tui@1.0.85.

Key Points

  • The @ai-sdk/provider/anthropic package now preserves native message batch request counts in provider metadata.
  • The Anthropic provider supports the full language-model option surface in batch requests.
  • Anthropic reasoning budgets are enabled for application inference profile ARNs within @ai-sdk/amazon-bedrock.
  • The @ai-sdk/amazon-bedrock package now accepts citation deltas in streaming responses.
  • A model family setting for embeddings has been added to Amazon Bedrock to support ARNs.
  • The AI SDK exposes individual image generation calls.
  • Parsed structured output is now exposed in streamText end callbacks.

Context

According to the Vercel AI SDK changelog, these updates include fixes and new features across various components of the SDK. For instance, the @ai-sdk/ai package addresses issues like summing Gateway image-generation costs across split requests and ensuring canonical hashes preserve undefined array element positions. The @ai-sdk/amazon-bedrock package also includes a change to omit assistant messages when only cache points remain after unsigned reasoning is filtered.

Why It Matters

These updates expand the capabilities and improve the reliability of the Vercel AI SDK for developers working with Anthropic and Amazon Bedrock models. The enhancements to batch request handling, streaming responses, and cost accounting can streamline development and deployment processes.

What To Do

  • Review the updated @ai-sdk/provider/anthropic documentation for changes to batch request handling.
  • Explore the new model family setting for embeddings in @ai-sdk/amazon-bedrock to support ARNs.
  • Note the exposure of individual image generation calls for potential integration into applications.
  • Test the streamText end callbacks for parsed structured output in applications.