OpenAI released GPT-Live-1 and GPT-Live-1 mini, a new generation of voice models built on a full-duplex architecture that lets ChatGPT listen and speak at the same time. The models can add short verbal cues like "mhmm," handle interruptions naturally, and decide many times per second whether to speak, pause, or invoke a tool.
GPT-Live-1 becomes the default voice model for paid ChatGPT tiers (Go, Plus, Pro), while GPT-Live-1 mini replaces Advanced Voice Mode for free users. Both models delegate heavier reasoning, search, and agentic work to GPT-5.5 in the background. Over 150 million people currently use ChatGPT voice features; an API version is planned soon. Video, screen sharing, and full multilingual parity are not yet available.
"Over time, we think this will unlock the ability to use voice as a primary interface to computing, and to manage increasingly complex long-running agentic work," said Atty Eleti, ChatGPT Voice Product Lead.