DeepMind launches Genie 4: real-time interactive world model with native audio
RubenEVillegas · x · 2026-10-28
Google DeepMind unveiled Genie 4, its most realistic and interactive world model yet: users can steer dynamic environments in real time (events, camera angles, actions), it generates native audio matched to the world—from background sounds to characters conversing live—and simulates photorealistic scenes with real-world physics and more coherent multi-step events.
Former Google researcher Ruben Villegas, who worked on it as his last project at Google, says its quality and dynamics still clearly outperform the 3D creations of the latest frontier models, and that bridging this gap toward accurate, efficient real-world simulation remains an open challenge key to useful physical intelligence.
Related event: DeepMind Unveils Genie 4, Its Most Realistic Interactive World Model(38 posts)→
More from Models
- Screenshot Appears to Show GPT-6 and GPT-5.6 Sol in ChatGPT Model Picker — ShelterCultural5725 · 2026-10-28
- ChatGPT Work sync bug makes messages vanish from view while the model still recalls them — LynxAirSound · 2026-10-28
- Palisade Research finds Alibaba's Qwen 3.8 models routed some requests to company servers — The Decoder · 2026-10-28
- DeepSeek on $75K of hardware: 2,120 tok/s decode at 16 concurrent sessions — victormustar · 2026-10-28
- Frontier models hit 73.3% on SpatialBench 10 months after release, near 75% ceiling — kenbwork · 2026-10-28
- 50-task Salesforce agent eval: GPT-6.1 Sol scores 72%, well behind Claude Opus at 87% — TruthDeep8659 · 2026-10-28