Google has announced Genie 4, a new world model that delivers enhanced realism and improved real-time interactive simulation. Genie 4 is natively multimodal, generating audio-like music, ambient noises, and interactive dialogue simultaneously with the simulated world. It builds on the Gemini family: Gemini Omni anchors simulations in real-world appearance and behavior, while other Gemini models provide the reasoning needed to keep a simulation coherent and continuous.
Users can set an event in motion and watch Genie 4 simulate what follows, including how the world responds, how it sounds, and how people in it move, react, and speak. Using natural language, users can modify a scene in real time, such as changing the weather, moving the camera, or introducing a new person or object. Gemini adapts each prompt to fit the surrounding scene and keeps the world progressing cohesively between inputs.
Genie 4 also advances physics and human action. It simulates scenarios such as aerodynamics and collisions, and it creates characters that move naturally, respond to their environments, and can be steered interactively. It is Google's first real-time world model to generate audio natively, from wind in the trees to characters who converse and react as events unfold.
Genie 4 is rolling out as a research preview for trusted testers, starting with Google AI Ultra subscribers in Project Genie and early access partners of the Gemini API and Google AI Studio. Google says understanding and simulating the world is a crucial part of building AGI, and that Genie 4 can help develop training environments for robots and agents as well as new interactive experiences. A waitlist is open for Project Genie.
Source: Google DeepMind Blog · Summarized by HeadlinesBriefing