Build more natural voice experiences with GPT‑Live‑1 in the API
GPT-Live-1 is now available in the API at $0.05 per minute for the front-end voice layer.
OpenAI says the model brings full-duplex voice conversations to developers, handling listening and speaking in one model instead of a stitched STT-LLM-TTS stack. The release emphasizes smoother interruptions, silence and background-noise handling, long-session reliability, telephony support, and prompt-level control over tone, pace, and style. GPT-Live-1 can also delegate deeper reasoning and tool calls to a backend model or third-party system. OpenAI cites early customer results from Speak, Yelp, Fin, Cognition, and another customer quote on reduced code complexity. OpenAI's note
OpenAI says the model brings full-duplex voice conversations to developers, handling listening and speaking in one model instead of a stitched STT-LLM-TTS stack. The release emphasizes smoother interruptions, silence and background-noise handling, long-session reliability, telephony support, and prompt-level control over tone, pace, and style. GPT-Live-1 can also delegate deeper reasoning and tool calls to a backend model or third-party system. OpenAI cites early customer results from Speak, Yelp, Fin, Cognition, and another customer quote on reduced code complexity. OpenAI's note
score 7