Google releases Genie 4 world model with native audio and real-time steering
Google announced Genie 4, a natively multimodal world model that generates real-time interactive simulations with audio. It is rolling out as a research preview for trusted testers, with wider access planned for Google AI Ultra subscribers and Gemini API partners.
Key facts
- Genie 4 is Google's first real-time world model to generate audio natively, including ambient sound and interactive dialogue.
- Gemini 3.8 Flash serves as a guiding agent, observing scenes and breaking goals into steps to maintain simulation coherence.
- Users can modify simulations in real time with natural language, such as changing weather, moving the camera, or adding objects.
- All video and audio generated by Genie 4 includes SynthID, Google's imperceptible digital watermarking technology.
- Google says visual details and object permanence can still drift over longer sessions or after multiple complex prompt edits.
Why it matters
Genie 4 combines Gemini reasoning with real-time pixel-level simulation, an early step toward hybrid world models Google describes as more coherent and controllable. Google positions world models as a component of AGI research and as training environments for robots and agents.