genie 4
Describe a setting and Genie 4 renders a playable, photoreal world — physics, ambience and character voices all respond to you live.
Feedback
AI Dance Tools
Explore our AI-powered dance video creation tools
MiniMax H3
MiniMax H3 AI Video Generator
Seedance 2.5
The Future of AI Video Is Here.
Seedance 2.0
The Future of AI Video Is Here.
Veo3.1
Create Stunning Videos with Veo3.1
FLUX 3 Video Generator
Kling 3.0
Next-Gen AI Video Generator
Grok Video Generator
Create Videos from Text or Images with AI
MiniMax H3 video generator
MiniMax H3 AI Video Generator
A Closer Look at the Genie 4 World Model
Feed Genie 4 a written idea or one reference frame and it returns a navigable scene where picture, motion, audio and voices are produced together.
- One Model, Every ModalityAudio, ambience and spoken lines are produced in the same pass as the imagery, rather than layered on after the Genie 4 scene already exists.
- Steer the Scene by TypingAsk for different weather, a fresh camera angle or an extra character mid-run, and Gemini re-fits the simulation to what is already on screen.
- Powered by the Gemini LineupGemini Omni keeps appearance and motion anchored to reality, while companion Gemini models track continuity from one frame to the next.
Four Steps to Your First Genie 4 Scene
Four short steps take you from the Genie 4 waitlist to a world you can steer in real time.
What Genie 4 Delivers in Real Time
Six strengths that define what this generation of Genie 4 world simulation delivers in real time.
Weight and Impact That Feel Real
Objects carry convincing mass and momentum, so airflow, collisions and contact behave the way you would expect.
Lifelike Character Movement
Figures walk, react and talk in a way that reads as human, and you can redirect them while the simulation keeps running.
Sound Born With the Picture
Each location brings its own soundscape — wind through leaves, surf breaking, distant traffic — generated alongside the visuals.
Conversation That Responds
Characters reply to your prompts and to events inside the world as they happen, and their speech shifts as the scene evolves.
Planning Across Many Steps
Gemini 3.8 Flash runs inside the loop, mapping out actions, reviewing frames and re-planning whenever the world takes a turn.
Cinematic, Photo-Grade Look
Light, grain and lens behavior hold steady, so long takes and free camera moves still read as real footage.
Questions About Genie 4
Short answers to the questions people ask most about the Genie 4 world model.
So what is Genie 4, in plain terms?
A multimodal world model that predicts and renders how a scene will look, sound and react next.
How can I try it?
Access is limited to vetted testers during the research preview — sign up for the Project Genie waitlist, the Gemini API or AI Studio.
Does it produce audio and voices too?
Yes. Ambience and character dialogue come from the model itself and shift with whatever you type.
What controls do I have inside a scene?
Keyboard input moves you through the space, and text prompts reshape the environment while it keeps running.
Where do the Gemini models fit in?
Gemini Omni anchors appearance and behavior; companion Gemini models keep the simulation consistent as time passes.
What can people build with it?
Robotics and agent training grounds, navigation benchmarks for AI, and a new class of interactive experiences built on Genie 4.
Get Early Access to Genie 4
Ask for a seat in Project Genie, or test Genie 4 sooner through the Gemini API and Google AI Studio.