# xAI Grok audio models now available on Vercel AI Gateway

> Source: <https://vercel.com/changelog/xai-grok-audio-models-now-available-on-vercel-ai-gateway>
> Published: 2026-06-29 00:00:00+00:00

[xAI's audio models](https://vercel.com/ai-gateway/models?capabilities=realtime&providers=xai) are now live on AI Gateway. Realtime voice, text to speech, and speech to text are all available through the [AI SDK](https://ai-sdk.dev/v7/docs/ai-sdk-core/realtime) with the same routing, observability, and spend controls as your other models.

These capabilities are available on the AI SDK 7 release.

Capability | Models |
|---|---|
|
|
|
|
|
|

A voice agent has two pieces: a server route that mints a short-lived token, so your API key never reaches the client, and a browser component that connects with it.

Add the token route: this example sets [model](https://vercel.com/ai-gateway/models/grok-voice-think-fast-1.0) to `xai/grok-voice-think-fast-1.0`

:

Then connect from the browser. The `useRealtime`

hook from `@ai-sdk/react`

fetches that route and manages the WebSocket connection, microphone capture, and audio playback:

Generate [spoken audio](https://vercel.com/ai-gateway/models/grok-tts) from text with `generateSpeech`

. Pass a voice and an output format, then write the result to a file with `xai/grok-tts`

:

[Transcribe](https://vercel.com/ai-gateway/models/grok-stt) recordings into text with `transcribe`

. This example uses `xai/grok-stt`

:

You can also try the xAI audio models directly in the AI Gateway playground. Open the [models list](https://vercel.com/ai-gateway/models) and click into any of the models to use them directly in the browser. The `xai/grok-voice-think-fast-1.0`

playground [here](https://vercel.com/ai-gateway/models/grok-voice-think-fast-1.0) allows you to talk to the agent and see responses instantly: