Models · Updated 28 Oct, 04:51 pm IST
Google DeepMind unveils Genie 4 world model with audio and interactive visuals
Why it matters for readers: It demonstrates how AI can create convincing, changeable virtual scenes with matching audio and dialogue in real time.
- Genie 4 can create photorealistic visuals and interactive environments that users can change mid-simulation using text prompts.1
- The model is trained to generate real-time audio that aligns with the simulated world, including music, ambient sounds, and interactive dialogue.1
- Genie 4 was released as a research preview on Tuesday, October 27.1
- Genie 4 succeeds Genie 3 and enters a small but growing field of world models that predict physics and object interactions.1
- Other recent activity in the world-model space includes Runway's GWM Worlds 2 and industry consolidation such as AMD’s acquisition of World Labs.1
Get a brief like this every morning
Uzha reads hundreds of sources and gives you the stories that matter for your work, with every source linked. Free.
Get started