Introducing Gemini 3.8 Live and 3.8 Live Extended Thinking
Google launches two voice-focused models, Gemini 3.8 Live and Live Extended Thinking, to make spoken conversations with AI feel more natural and better suited to complex tasks. The standard model stresses scale, cost efficiency and visual understanding, while Extended Thinking handles harder multi-step reasoning; both roll out via the Gemini API, Google AI Studio, Search Live, Gemini Enterprise an
What the article says
On September 15, 2026, Google announced Gemini 3.8 Live and Gemini 3.8 Live Extended Thinking, described as its most advanced live dialogue models, designed for voice agents and natural conversation. Gemini 3.8 Live targets scale and cost efficiency with visual grounding; Extended Thinking targets complex tasks requiring multi-step reasoning.
3.8 Live processes visual input in near real time, switches between 97 supported languages mid-conversation and runs tools and API calls in the background while it keeps talking. Extended Thinking reasons while speaking, uses cues such as "Let me check that…" and narrates progress on multi-step tasks. For Extended Thinking, Google cites first place on Artificial Analysis' Speech to Speech Quality Index (82.6), 68.6% on τ-Voice and 97.7% on Big Bench Audio. Rollout is staggered: the Gemini API and AI Studio, private preview for enterprises, Search Live, Gemini Live and some Workspace products. All audio carries a SynthID watermark.
Read the official article AnnuIA profile
Summary written by AnnuIA in its own words from the official page of Google DeepMind. The content and rights belong to the publisher: please read the original article.