Mobile & Apps

AI, Mobile & Apps

OpenAI Launches GPT-Live-1 Voice Models With Full-Duplex Conversation and Live Translation

OpenAI launched GPT-Live-1 and GPT-Live-1 mini on July 8, 2026 with full-duplex voice that can process incoming speech while generating replies, supporting mid-sentence interrupts and live translation. GPT-Live-1 mini becomes ChatGPT’s default, while paid users get GPT-Live-1. Atty Eleti said he’s used it for 30–40 minute conversations; 150M already use voice/dictation.

Scroll to Top