On July 8, 2026, OpenAI launched two full-duplex voice models in ChatGPT: GPT-Live-1, the new default for paid users, and GPT-Live-1 mini, the default for free users. Both replace Advanced Voice Mode (source: OpenAI official announcement, 2026-07-08).
Full-duplex means the model listens and speaks at the same time. Advanced Voice Mode relied on turn-based silence detection — it waited for you to stop talking before it responded. With GPT-Live, you can interrupt mid-sentence, and the model can start answering while you are still speaking.
According to OpenAI, GPT-Live delegates harder questions to GPT-5.5 behind the scenes, while the voice interaction itself stays with GPT-Live. A system card covering both models, dated July 8, 2026, was published alongside the release.
Source: OpenAI official announcement, 2026-07-08
