Google launches Gemini 3.8 Live voice models with live thinking
Google introduced Gemini 3.8 Live and 3.8 Live Extended Thinking, voice-focused models that reason while they speak, switch between 97 languages and run tools in the background.
Google's most advanced live dialogue models
On September 15, 2026, Google introduced Gemini 3.8 Live and Gemini 3.8 Live Extended Thinking, two voice-first models it calls its most advanced live dialogue models yet. Both are available today through the Gemini API, Google Workspace, Search and the Gemini app, aimed at developers and enterprises building production-ready voice agents.
Talk and work at the same time
Gemini 3.8 Live is tuned for scale and cost efficiency: it processes visual input in near real time, switches automatically between 97 languages mid-conversation and runs tools and API calls in the background without interrupting the dialogue.
Gemini 3.8 Live Extended Thinking targets complex, multi-step tasks. It reasons and speaks at the same time, acknowledges requests with cues like “Let me check that…” and narrates progress live. Google reports first place on Artificial Analysis's Speech to Speech Quality Index at 82.6 points and 97.7% on Big Bench Audio.