Sign in
← AI PulseAug 6, 2026

Wire · news · Multi-source brief

Vercel AI SDK Updates Include Fish Audio Provider and Enhanced Instruction Handling

Vercel has released updates to its AI SDK, introducing a new Fish Audio provider for speech and transcription models, alongside improvements in how language model instructions are managed and applied.

By Illumora Editorial

Source · Aug 6, 2026, 12:58 AM · On Illumora · Aug 6, 2026, 1:08 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/voyage@2.0.22: Version Packages (#18484) · vercel/ai
Save

Vercel has updated its AI SDK, with releases including @ai-sdk/voyage@2.0.22, @ai-sdk/vue@4.0.54, @ai-sdk/vercel@3.0.24, and @ai-sdk/tui@1.0.55. These updates introduce a new audio provider and refine instruction handling for language models.

Key Points

  • The @ai-sdk/fish-audio@3.0.0 package adds a new provider for speech and transcription models.
  • ai@7.0.54 allows ToolLoopAgent prepareCall callbacks to read and override top-level reasoning options.
  • A defaultInstructionsMiddleware has been added to ai@7.0.54 for applying default language model instructions while preserving call-level overrides.
  • ai@7.0.54 now preserves preceding assistant messages when regenerating a response.
  • @ai-sdk/anthropic@4.0.32 rejects spliced Anthropic generations.
  • @ai-sdk/anthropic@4.0.32 allows duplicate message start events for the active message.

Context

According to the Vercel AI SDK Changelog, these updates are part of a series of package releases. The changes are designed to enhance the SDK's capabilities, particularly in audio processing and the nuanced control of language model behavior. The Changesets release GitHub action manages these updates, publishing packages to npm automatically upon merging.

Why It Matters

These updates offer developers more granular control over AI model interactions and expand the SDK's multimodal capabilities. The introduction of a dedicated audio provider and improved instruction management can lead to more robust and flexible AI applications, particularly for those integrating speech and transcription functionalities or requiring precise control over model responses.

What To Do

  • Review the @ai-sdk/fish-audio@3.0.0 documentation to explore the new speech and transcription models.
  • Investigate the ToolLoopAgent prepareCall callbacks in ai@7.0.54 for custom reasoning overrides.
  • Test the defaultInstructionsMiddleware in ai@7.0.54 to manage language model instructions.
  • Note the changes in @ai-sdk/anthropic@4.0.32 regarding spliced generations and message start events.