Google DeepMind launches Genie 4, a real-time interactive world model with native audio
rakyll · x · 2026-10-28
Google DeepMind has unveiled Genie 4, its most realistic and interactive world model yet. Key capabilities:
- Real-time steerability: users shape dynamic environments, events, camera angles and actions live
- Native audio: from realistic background sounds to characters conversing in real time
- Photorealistic scenes with simulated real-world physics
- Coherent multi-step events via Gemini reasoning; dev Elvis Collins says it's built on Gemini Omni with Gemini in the loop for coherence
Rolling out as a research preview to trusted testers, coming to Google AI Ultra subscribers via Project Genie and early-access partners through the Gemini API and AI Studio.
Related event: DeepMind Launches Genie 4, Its Most Realistic Interactive World Model(41 posts)→
More from Multimodal
- Stunning AI-generated concept art casts goddess Sekhmet walking a post-apocalyptic wasteland — DionysianAgent · 2026-10-28
- Custom node pack brings looping H3 video generation to Comfy Cloud — PurzBeats · 2026-10-28
- Reproducible YuE2 pipeline: AI-generated 1980s Eurodisco song with full settings — Robonotes1760 · 2026-10-28
- NVIDIA ships 2,575 sim-ready 3D assets on Hugging Face as Polyfork opens 2,600+ free models — Vjeux · 2026-10-28
- Multi-model comparison finds some AI models just render people as hotter — r13i · 2026-10-28
- Hands-on with SpAItial Echo-3: record-shop scan turned virtual experience, pricing detailed — risingodegua · 2026-10-28