Gemini 3.1 Flash Live: Making audio AI more natural and reliable

Gemini 3.1 Flash Live is the latest high-quality audio and voice model built for real-time dialogue. Developers can access it in preview via the Gemini Live API in AI Studio, enterprises through Gemini Enterprise for Customer Experience, and everyone via Search Live and Gemini Live, which now operate in more than 200 countries and territories.

Generative AI is experimental. The model improves reliability for voice-first agents and task execution at scale. On ComplexFuncBench Audio it leads with a score of 90.8% compared with the previous model, and it tops Scale AI’s Audio MultiChallenge with a score of 36.1% when “thinking” is on.

Those benchmarks measure multi-step function calling, complex instruction following and long-horizon reasoning amid the interruptions and hesitations typical of real-world audio. Tonal understanding has been enhanced to produce more natural dialogue, and the model better recognizes acoustic nuances such as pitch and pace than 2.5 Flash Native Audio.

gemini 3.1, gemini live, flash live, audio ai, voice model, real-time dialogue, ai studio, gemini enterprise, complexfuncbench, audio multichallenge