Genie 4 turns a sentence into a playable world you can walk through
This project is scheduled for launch
Launch date: Tuesday, February 23, 2027 at 08:00 AM UTC

Genie 4 is Google DeepMind's natively multimodal world model, announced October 27, 2026. Instead of rendering a fixed clip, it builds a playable, physics-aware environment you can steer in real time with plain language, complete with ambient audio and characters that answer you.
The genie4 guide is an independent walkthrough of the announcement: what Genie 4 actually does, how prompting an interactive world differs from prompting a video, a side-by-side comparison with Genie 3, plus the staged access picture (trusted testers first, then AI Ultra via Project Genie, Gemini API and AI Studio partners).
Highlights worth knowing: native multimodality across image, audio and dialogue; real-time steering of weather, camera and characters mid-session; Gemini Omni grounding with a Gemini 3.8 Flash agent loop for long-horizon consistency; and SynthID watermarking on every generated frame and audio track. Google has not published frame rate, resolution or pricing, so the guide deliberately leaves those blank rather than guessing.
Comments will be available once the project is launched.
Visit Website
genie4.art
Don Wilson
@don495627
Share this launch.