Google DeepMind unveils Genie 4, its new world simulation model
Google DeepMind announces Genie 4, a multimodal world model simulating interactive environments in real time with native audio, first offered to trusted testers.
By GennnAi··1 min read
Genie 4 also generates music, ambient sounds and interactive dialogue. Gemini Omni anchors the simulations in real-world appearance and behavior. Users modify the scene in natural language, such as changing the weather or moving the camera. Google sees a tool for training robots and agents. A research preview is rolling out to trusted testers, then to Google AI Ultra subscribers in Project Genie.
Source: Google DeepMind, published on October 27, 2026.