Spatial intelligence
Spatial intelligence AI connects generated scenes, camera motion, memory, interaction, and physical logic around one focused concept.
Spatial interfaceScene explainer
Each page reads as a visual path first, then keeps the source-backed links nearby.
01Category shift
A clip can look amazing and still be over when the camera stops. A world model has to keep the place coherent when the viewer moves, acts, or returns.
World model vs video model
02Why now
Marble makes generated 3D spaces feel like a product surface. Genie and Oasis push toward interactive worlds. HY-World 2.0 and LingBot-Map show reconstruction and persistent assets.
Browse model dossiers
03Reader path
The fastest way in is visual: watch a generated place, understand what changed, then choose whether to open models, comparisons, or the timeline.
Category shift
A clip can look amazing and still be over when the camera stops. A world model has to keep the place coherent when the viewer moves, acts, or returns.
Spatial intelligence is the layer that connects perception, imagination, motion, and consequence inside one environment.
Why now
Marble makes generated 3D spaces feel like a product surface. Genie and Oasis push toward interactive worlds. HY-World 2.0 and LingBot-Map show reconstruction and persistent assets.
Project Sid adds another layer: once a world persists, agents can live inside it, organize it, and make the scene feel social.
Reader path
The fastest way in is visual: watch a generated place, understand what changed, then choose whether to open models, comparisons, or the timeline.