AI · AI model
Project Genie
Prototype web app generating playable, interactive worlds in real time.
Project Genie is an experimental research prototype from Google DeepMind that lets users create, explore, and remix interactive virtual environments. Powered by the Genie 3 world model alongside Nano Banana Pro and Gemini, it generates the path and environment ahead in real time as you navigate. It is initially available to Google AI Ultra subscribers in the United States.
What it's for
- Generate playable 3D environments from text prompts or uploaded images
- Preview and tweak initial scenes and character perspectives using visual sketching
- Navigate interactive virtual worlds with responsive real-time generation
- Remix community-generated worlds and export gameplay video clips
Features
- Real-time environment generation powered by the Genie 3 world model
- World Sketching image preview and editing powered by Nano Banana Pro
- Multi-perspective navigation supporting first-person and third-person views
- Dynamic traversal modes including walking, driving, and flying
- World remixing and community inspiration gallery
- Video recording and export for exploration sessions
- 60-second generation session limits
Pricing
Paid. Requires a Google AI Ultra subscription; US users 18+ only during prototype rollout
Details
- Type
- AI model
- Category
- AI
- Topic
- World Models
- Open source
- No
- Platforms
- Web
- For
- AI researchers, game designers, and generative media creators
- Alternative to
- Oasis (Decart), GameNGen
- Tags
- world models, generative ai, google deepmind, interactive ai, gaming, genie
Why it was saved
- @GoogleDeepMind: First public access to Genie 3 world models as a consumer-facing product; shows how world-model tech ships as an actual playable interface.
Related
Spatial AI models that turn images, text, and video into 3D worlds
AI creative suite for generating and editing video, images, and animation
AI platform generating and editing video, image, and synchronized audio.
AI video and image generator known for cinematic motion and physics
Generative AI video model producing cinematic footage with native audio
AI video and image generation from text prompts and static images
Updated