Google launches Genie 4 world model with audio and real‑time editing
Genie 4, a research‑preview world model, adds photorealistic visuals, real‑time audio and live prompt‑based edits, expanding to two‑minute simulations.
Google DeepMind unveiled Genie 4 as a research preview on October 27, 2026. The model generates photorealistic, interactive worlds and includes real‑time audio such as music, ambient sounds and dialogue.
Users can modify simulations mid‑run with natural‑language prompts, changing weather, camera angles, or adding objects and characters. Simulations now last up to two minutes, double the duration of Genie 3.
Genie 4 runs on Gemini 3.8 Flash, which coordinates the visual generation while Gemini acts as a director.
Google pitches Genie 4 for pre‑visualization in film, game prototyping and robot training scenarios. The company notes that longer runs may suffer detail drift and audio accuracy challenges.
The release follows recent industry moves, including Runway’s GWM Worlds 2 and acquisitions such as AMD buying World Labs. Google positions Genie 4 as a step toward hybrid world models that blend text generation and real‑time simulation.
What it means for you
Genie 4 offers creators a more immersive, editable AI‑generated environment, potentially speeding up storyboarding and robotics testing. For users preferring free or locally hosted models, the research‑preview remains cloud‑based, highlighting the gap between cutting‑edge proprietary tools and open‑source alternatives.
On FreeLLMArena
From our rankings on 27 Oct 2026, based on Arena's blind votes:
- Gemini 3.8 Flash (High): #9 of 418 in Chat, paid
Sources
- Google’s Genie 4 world model adds audio, more realistic worlds, and real-time steering, TechCrunch, 27 Oct 2026
- Google’s next Genie AI model generates more realistic experiences, The Verge, 27 Oct 2026
Written automatically from the sources above, then checked by our code: every number and name must appear in them, and no sentence may copy a source. Spotted a mistake? Report it.
Which AI can you use for free today?
Every free AI model, ranked by blind human votes, with the memory and speed it needs on your own GPU.