From the source
Google DeepMind is announcing Genie 4, a world model that generates real-time interactive simulations with native audio, and has begun rolling it out as a research preview for trusted testers.
- The model is natively multimodal, so music, ambient noise and interactive dialogue are generated at the same time as the world itself, and all the sounds in the demos come directly from Genie 4.
- Genie 4 builds on capabilities from the Gemini family: Gemini Omni anchors simulations in real-world appearance and behavior, while other Gemini models supply the reasoning that keeps a simulation coherent and continuous.
- Users steer scenes with standard keyboard controls and text prompts, and can change the weather, move the camera or add a person or object in real time while Gemini adapts the prompt to the surrounding scene.
- DeepMind says Genie 4 sets a new standard of realism for real-time interactive simulation, built on physical understanding that covers aerodynamics and collisions, with the biggest leap in how it simulates people.
- The research preview is limited to trusted testers for now, with a wider release planned for consumers and developers, starting with Google AI Ultra subscribers in Project Genie and early access partners of the Gemini API and Google AI Studio.
