Models · Updated 27 Oct, 10:55 pm IST

Google DeepMind announces Genie 4 world model with real‑time audio and steering

Why it matters for readers: It lets you imagine and later try more realistic, playable AI‑generated scenes with sound and talking characters.

Genie 4 is a natively multimodal world model that can generate visuals, ambient audio, music, and interactive dialogue in real time.1 2 3 4
The model lets users update scenes mid‑simulation with natural language prompts and supports more realistic physics, character movement, and object interactions.1 2 3 4
Genie 4 can maintain consistent simulations for about two minutes, doubling the duration reported for the prior Genie 3 research preview.1 3
Genie 4 builds on and integrates Gemini family models (including Gemini Omni and Gemini 3.8 Flash) to anchor appearance and provide planning and reasoning for coherence.1 2 4
Google has started a limited research preview with trusted testers and opened a waitlist for Project Genie and Google AI Ultra subscribers.1 2 3

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.

27 Oct, 09:30 pm · The Verge AI
Google’s next Genie AI model generates more realistic experiences
27 Oct, 09:31 pm · Google DeepMind
Genie 4: A new generation of world simulation
27 Oct, 10:00 pm · TechCrunch AI
Google’s Genie 4 world model adds audio, more realistic worlds, and real-time steering
27 Oct, 10:55 pm · The Decoder
Google Deepmind's Genie 4 puts a modern spin on agent designs that predate LLMs