From the source
Google Deepmind is releasing Genie 4, which it calls its most advanced world model for generating interactive 3D worlds, and opening it to Google AI Ultra subscribers.
- The model was previously available only as a research preview, and the release is part of a major update to Project Genie.
- Genie 4 lets users reshape worlds through text input in real time, changing weather, characters or storylines, and it generates native real-time sound that adapts to the scene.
- The older Genie 3 also gets longer sessions of up to three minutes, 360-degree panorama support and chainable worlds.
- A new AI agent builds worlds from spoken descriptions, and a community gallery lets users share and remix creations.
- All generated content carries SynthID watermarks and goes through safety checks, while Google AI Pro members can join a waitlist.