Gemini 3.1 Flash Live
Gemini 3.1 Flash Live is Google’s highest-quality real-time audio model, released March 26, 2026. It delivers natural, low-latency voice conversations with improved tone recognition — better discerning pitch, pace, and nuance — and filters out background noise like traffic or TV. It supports over 90 languages and doubles the conversation context length of the previous model, scoring 90.8% on ComplexFuncBench Audio for multi-step function calling. Available via the Gemini Live API in Google AI Studio, powering Search Live globally and the Gemini app across 200+ countries.
Visit
