OpenAI has launched GPT-Live-1 in the API, providing developers with a single model capable of simultaneous listening and speaking to build voice-enabled applications. This release aims to simplify voice layer development by replacing traditional chained architectures with a unified approach that handles interruptions and context more naturally.
Key capabilities include improved interruption handling via joint reasoning over incoming and outgoing audio, and the ability to delegate deeper reasoning tasks to backend models like GPT-6 Astra or third-party systems. The model supports tone, pace, and style customization through system prompts, manages background noise and silence effectively, and ensures reliability across long sessions.
Evaluations show GPT-Live-1 improves Full Duplex Bench performance by 30 percentage points over GPT-Realtime-2.1 and ranks #1 on the Tau3 benchmark for voice-agent intelligence when paired with GPT-6 Astra. It is available today at $0.05 per minute for the front-end voice layer, with expanded voice options and language availability planned for the coming months.