Google launches Gemini 3.8 Live voice models for production agents
Google has launched Gemini 3.8 Live and Gemini 3.8 Live Extended Thinking, two native speech-to-speech models designed for real-time voice agents. Both are available through the Gemini Live API and Google AI Studio for API-based production use. They are hosted models, so they cannot be self-hosted.
Gemini 3.8 Live is aimed at scalable, cost-efficient conversations with visual grounding, while the Extended Thinking version targets complex tasks requiring multi-step reasoning. The models can run tool and API calls in the background while audio continues streaming, process live visual input, parse alphanumeric information, switch among 97 languages during a conversation and combine audio with structured data. Extended Thinking can reason and speak simultaneously while narrating progress during longer tasks.
On Artificial Analysis’ Speech to Speech Quality Index, Gemini 3.8 Live Extended Thinking ranked first with 82.6. It recorded 68.6% on τ-Voice, 35.1% on Sierra’s τ-Voice-banking benchmark and 97.7% on Big Bench Audio. Both models cost $0.005 per minute for audio input and $0.018 per minute for audio output. Generated audio includes an imperceptible Google DeepMind SynthID watermark.
Written by Sakhanda Wire AI from reporting by MarkTechPost. Facts, quotes and full context are in the original article below.