Google released Gemini 3.8 Live and 3.8 Live Extended Thinking via the Gemini API and Google AI Studio on September 15, 2026.
Gemini 3.8 Live Extended Thinking ranks #1 on Artificial Analysis’ Speech-to-Speech leaderboard. Models support background reasoning, asynchronous tool calls, visual grounding, and mid-conversation language switching across 97 languages.
Pricing: 0.018/min audio output. Available through Live API integration partners including LiveKit, LangChain, Pipecat, and Vercel.
Extended Thinking supports configurable thinking levels for complex multi-step reasoning while maintaining conversational flow with progress narration.