Google Previews Genie 4 3D World Model

Google previewed Genie 4, an AI model that generates realistic three-dimensional environments with fluid movement, complex details, and accompanying audio. Users create and edit virtual worlds using natural language prompts or keyboard controls. The model relies on Gemini Omni for visual realism and Gemini 3.8 Flash for frame consistency. Google DeepMind is also exploring robotics applications, pairing Genie 4 with Gemini 4 Argon to simulate robots from high-level instructions. Google plans to release Genie 4 through its AI Ultra subscription, the Gemini API, and Google AI Studio after a limited early-access period.