Google Releases Live Avatar Feature for Gemini Conversational AI

Google has released Gemini 3.8 Live with Live Avatar, a feature that generates near real-time video for its conversational AI models. This visual presence layer is designed to enhance virtual offerings such as customer service and interactive walkthroughs.

The release follows the introduction of Gemini 3.8 Live on September 15, 2026, which was positioned for scalable, cost-efficient conversation and high-complexity reasoning tasks. The new feature is available in Gemini Enterprise with endpoints in the United States and European Union.

According to Google, the precise lip-syncing, natural expressions, and fluid turn-taking of Live Avatar allow enterprises to expand their virtual offerings. The technology has been previewed at Google Cloud Next 2026 and is generally available as of September 24, 2026.