Grok Voice Think Fast 2.0 now available on AI Gateway
By Steven Van ·
xAI's speech-to-speech model reasons while talking, fires tool calls sooner, and holds up to background noise and telephony compression.
Grok Voice Think Fast 2.0, a speech-to-speech voice model from xAI, is now available on Vercel's AI Gateway. It improves on the earlier Grok Voice model in reasoning, transcription accuracy, and conversation, and it reasons in parallel with speech so it can think through a query while talking without adding latency.
The model has also been trained to use fewer reasoning tokens than before, so tool calls tend to fire sooner, often before the end of the agent's first sentence, and its transcription holds up in real-world conditions like background noise and telephony compression. It's available as xai/grok-voice-think-fast-2.0 through the AI SDK's realtime API; developers mint a short-lived token on the server so the API key never reaches the client. See the realtime documentation to build a voice agent with it, or try the model in the AI Gateway playground.