Google DeepMind unveils Genie 4 world model with real-time audio and scene editing

Google DeepMind released Genie 4, its newest world model, in research preview on October 27, 2026. The model generates photorealistic interactive worlds and allows users to type prompts mid-simulation to alter the scene, such as changing weather, camera angles, or adding new people and objects.

For the first time, Genie 4 includes real-time audio generation, covering music, ambient sounds, and interactive dialogue. It maintains scene consistency for up to two minutes, double the limit of its predecessor, and offers improved physics and more human-like characters.

The model is closely integrated with Google's Gemini AI, which acts as a director by planning steps and checking frames while Genie creates the visuals. In a research experiment tied to Gemini Robotics, a Gemini 4 Argon model broke tasks into steps that Genie 4 simulated in real time, suggesting a possible way to test robot scenarios in simulation before real-world deployment.

Google said future hybrid world models could combine Gemini's text and code generation with Genie's real-time simulation. Potential use cases include gaming, interactive learning, and robotics. Genie 4 is available now as a research preview for trusted testers, with rollout to Google AI Ultra subscribers and early access partners of the Gemini API and Google AI Studio in the coming weeks.