Project Astra is Google DeepMind's research initiative toward a universal AI assistant that processes continuous video, audio, and text streams simultaneously with sub-300ms latency. Built on Gemini 2.5 Pro, it can reason about physical environments seen through a camera, navigate Android autonomously, and maintain persistent memory of past conversations and preferences. At Google I/O 2026, expanded rollout was announced via the Gemini API and Google AI Studio. It is not a standalone commercial product but a capability layer being integrated into Google products. Key features: - Real-time multimodal processing of unified video, audio, and text streams - Sub-300ms response latency via continuous stream processing (not discrete frame analysis) - Persistent memory of user preferences and past conversations - Autonomous Android OS navigation and task execution - Available to developers via Gemini API and Google AI Studio - Spatial understanding and physical-environment reasoning from live camera feed
Not directly priced as a standalone product. Accessible through Gemini API (pay-per-token) and Google AI Studio (free tier available).