Geni3 is an AI-powered 3D world generator that creates interactive, navigable environments in real-time from text or image prompts. Geni3 uses Google's Genie 3 model to generate worlds at 24 FPS with 720p resolution, maintaining environmental consistency for several minutes.
What is Geni3?
Geni3 is a web-based platform that transforms text descriptions or uploaded images into explorable 3D worlds in real-time. Powered by Google's Genie 3, it takes a prompt (up to 800 characters) and optional image input, outputting a playable 3D environment at 24 frames per second with 720p resolution. The platform is developed by Geni3 and runs in modern browsers.
Key Features
- Real-time generation at 24 FPS — Worlds are created and updated at 24 frames per second, allowing immediate exploration.
- 720p HD resolution — Output is rendered at 720p, providing crisp and detailed environments.
- Promptable world events — Users can modify worlds dynamically with text commands (e.g., change weather, add objects) while maintaining physical consistency.
- Visual memory and consistency — Geni3 remembers previously generated areas and keeps environments coherent for several minutes of exploration.
- Text-to-world and image-to-world — Supports both text prompts (up to 800 characters) and image uploads as input.
- Aspect ratio and duration options — Choose from 16:9, 9:16, 4:3, 3:4, 21:9, 1:1, and set video duration to 5 or 10 seconds.
- Interactive navigation — Generated worlds are fully navigable in real-time with responsive controls.
Who is it for?
- AI researchers and robotics engineers — Use Geni3 to generate unlimited training environments for autonomous agents, with consistent physics and promptable modifications for edge-case testing.
