Google launches gemini 3.8 live and extended thinking for enhanced voice interactions

Google has introduced Gemini 3.8 Live and Gemini 3.8 Live Extended Thinking, two advanced AI models designed to improve voice interactions. These models support real-time reasoning, multilingual capabilities, and background task execution, making conversations with AI more natural and intuitive. Gemini 3.8 Live Extended Thinking leads in task completion with 68.6% on τ-Voice and 35.1% on Sierra’s τ-Voice-banking benchmark, while scoring 97.7% on Big Bench Audio. Gemini 3.8 Live ranks second in the Speech Agent Arena and is noted for its cost-effectiveness. The models are available through the Gemini API, Google Workspace, and the Gemini app, enabling developers and enterprises to build production-ready voice agents. Partners such as Salesforce, Genspark, and Lumeris have expressed interest in the models due to their low latency and robust tool-calling capabilities. Audio generated by these models is watermarked with SynthID to ensure detectability. Gemini 3.8 Live Extended Thinking is rolling out today, with sign-up options for product updates and newsletters.