Sign in
← AI PulseJul 27, 2026

Wire · news · Multi-source brief

Vercel AI SDK Updates Mistral and TogetherAI Integrations

Vercel has released updates for its AI SDK, addressing issues in the Mistral integration and enhancing token usage reporting for TogetherAI.

By Illumora Editorial

Source · Jul 27, 2026, 9:24 PM · On Illumora · Jul 27, 2026, 9:27 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/mistral@3.0.52 · vercel/ai
Save

Vercel's AI SDK has received updates for its integrations with Mistral and TogetherAI, according to changelog entries published on July 27.

The @ai-sdk/mistral@3.0.52 release includes a fix to preserve reasoning in multi-turn conversations. Concurrently, the @ai-sdk/togetherai@2.0.68 update enables includeUsage for TogetherAI, allowing streaming responses to report token usage.

Key Points

  • The @ai-sdk/mistral@3.0.52 package was released on July 27.
  • This Mistral update fixes an issue with preserving reasoning in multi-turn conversations.
  • The @ai-sdk/togetherai@2.0.68 package was also released on July 27.
  • This TogetherAI update enables the includeUsage feature.
  • Enabling includeUsage allows streaming responses to report token usage.

Context

According to Vercel's AI SDK changelog, the Mistral integration previously had a bug that prevented the preservation of reasoning across multiple turns in a conversation. The TogetherAI integration, as noted in its changelog, lacked a mechanism to report token usage during streaming responses.

Why It Matters

These updates improve the reliability and observability of Mistral and TogetherAI models when used through the Vercel AI SDK. Developers can now expect more consistent conversational flow with Mistral and gain better insight into resource consumption with TogetherAI.

What To Do

  • Review the changelog for @ai-sdk/mistral@3.0.52 to understand the reasoning preservation fix.
  • Update to @ai-sdk/mistral@3.0.52 if using Mistral models in multi-turn conversations.
  • Check the changelog for @ai-sdk/togetherai@2.0.68 regarding the includeUsage feature.
  • Implement the includeUsage option for TogetherAI streaming responses to monitor token consumption.

Keep Exploring

/atlas/**mistral**-family /atlas/together-ai /studio?pack=foundation