ai@7.0.143
Vercel AI SDK 7.0.143 adds an experimental lowMemory option to streamText, defers chat ID generation in useChat() for prerendering, and introduces a box decision question type for TypeSafe AI.
Why this signal matters
Vercel AI SDK release 7.0.143 includes several patch changes: 1) streamText adds an experimental lowMemory: true option designed for single-consumer streaming without maintaining a replay buffer or retained final-content results, thereby lowering memory retention. 2) useChat() now generates default chat IDs on first access rather than immediately, allowing prerendering without reading random values when the ID is not used in rendering. 3) Added a box decision question type with image bounds for TypeSafe AI. 4) Dependency bumps for @ai-sdk/provider@4.0.29 , @ai-sdk/provider-utils@5.0.62 , and @ai-sdk/gateway@4.0.116 .
Actionable summary
ai@7.0.143 patch release introduces experimental lowMemory: true flag in streamText to avoid replay buffers and final-content retention for single consumers, defers default chat ID generation on first access in useChat() to support prerendering, and updates provider dependencies.
- Agent usefulness
- 77/100
- Confidence
- 96%
- Canonical data
- JSON + Markdown
What builders should check
- Upgrade to ai@7.0.143 to benefit from useChat prerendering fixes.
- Optionally test experimental: { lowMemory: true } (or lowMemory: true ) in streamText if memory usage during single-consumer streaming is a bottleneck.