AI SDK 7.0.150 released
A patch release adding experimental streaming speech generation from text deltas and parts.
What changed in 7.0.150
- New experimental_streamSpeech function streams audio from text using speech models with optional voice and provider options.
- Returns audio stream plus usage, warnings, responses and provider metadata; errors handled via onError callback.
- New types exported for speech streaming including Experimental_SpeechTextPart, Experimental_SpeechStreamPart, and Experimental_StreamSpeechResult.
- Mock streaming speech model added to ai/test for testing purposes.
Summarised by AgentGid from Vercel's official release notes on GitHub, as reported there; read the original before relying on it. Price, status and release dates come from our daily data.
AI SDK at a glance
| What it is | Vercel's TypeScript toolkit for building AI applications and agents, with one API across model providers and UI helpers for streaming. |
|---|---|
| Latest version | 7.0.179, Nov 20, 2026. This page is about an earlier release. |
| Release pace | 63 stable releases in the last 30 days; 7.0.150 came the same day as the previous one |
| Price | Free (OSS). AI SDK is open source and free to use; you pay only for the models and infrastructure it runs on. (checked Nov 21, 2026) |
| Downloads, last 7 days | 48.7M up 8.6% |
| Gid Score | 67 #3 in Agent frameworks and SDKs |