Genie 4 is a browser-based AI tool that generates interactive 3D worlds from text prompts in real time at 24 FPS and 720p resolution, powered by Google DeepMind's Genie 4 World Model.
What is Genie 4?
Genie 4 is a text-to-3D world generation platform from Google DeepMind that turns a short text description (or an uploaded image) into an explorable, photorealistic 3D scene. The world runs in the browser at 24 frames per second and 720p resolution, with consistent physics and layout for several minutes. Users can navigate the environment, modify it with additional text prompts, and export captures or playable builds—no 3D modeling experience or installation required.
Key Features
- Real-time 3D generation — generate interactive worlds from text at 24 FPS and 720p, with physics and scene coherence lasting several minutes.
- Text and image input — create worlds using a text prompt (up to 800 characters) or upload a starting image to transform into a playable 3D environment.
- Promptable editing — change weather, lighting, or add objects by typing new commands; the world updates instantly while maintaining consistency.
- Aspect ratio and duration options — choose from 16:9, 9:16, 4:3, 3:4, 21:9, or 1:1 for export captures, and set video duration to 5 or 10 seconds.
- Commercial license — included in all paid plans; generated content can be used commercially without a watermark.
- Multi-device support — runs in modern browsers on desktop, tablet, and phone (performance varies by device).
Who is it for?
- AI researchers — train reinforcement learning agents in diverse, on-demand environments with stable physics for minutes.
- Educators — create immersive 3D scenes for history, geography, or science lessons that students can explore in real time.
- Creative professionals — rapidly prototype game environments, animated scenes, or fictional settings from text descriptions.
- Robotics engineers — stress-test models with novel environments without building custom simulations from scratch.
Use cases
- Research and simulation — generate consistent environments for training agents at scale without manual level design.
- Immersive learning — allow students to step inside abstract concepts like historical events or biological ecosystems.
- Creative world-building — craft fantasy worlds or product visualization scenes from rough text ideas in minutes.
- Scientific visualization — build physical ecosystems or environmental scenarios for study and presentation.
How does Genie 4 work?
The process has four steps: (1) describe your world with a short text prompt or upload an image, (2) click generate to produce an interactive 3D scene, (3) explore and interact in real time—move around, view from any angle, (4) export a playable build or a short capture, or keep editing with new prompts.
Pricing
Genie 4 operates on a freemium model: new users receive free credits to start. Paid plans are billed annually or monthly: Basic ($17.9/month, annual, 12,000 credits/year), Professional ($29.9/month, annual, 24,000 credits/year), and Enterprise ($59.9/month, annual, 72,000 credits/year). All paid tiers include a commercial license, no watermark, and priority support. Cancel anytime.
FAQ
Is Genie 4 free to use?
Yes, new users get free credits to start. The free tier allows limited generations; paid plans unlock more credits and additional features.
What resolution and frame rate does it output?
Generated worlds run at 720p resolution and approximately 24 FPS in the browser. The environment remains consistent for several minutes of exploration.
Can I use Genie 4 for commercial projects?
Paid plans include a commercial license and no watermark. Free-tier users should review the terms for commercial usage rights.
Which devices does Genie 4 support?
It runs in modern browsers on desktop, tablet, and mobile devices. Performance may vary, especially on older phones.
What inputs can I use to create a world?
You can type a text prompt of up to 800 characters or upload an image to generate a 3D world. You can also combine both.
