PlayGen2 real-time interactive world model released
What happened
Jihai Zhang and others proposed PlayGen2, a real-time interactive world model that separates learned 3D state evolution from neural rendering, maintains persistent world memory using volumetric occupancy grids and a 3D state with identity indices, and has released its code and pretrained weights. On the Doom test platform, the model learned world dynamics that include geometry-aware player movement and player-NPC interactions; the authors report that the full simulation and rendering pipeline has a backend latency of about 160ms on a MacBook Pro with an M4 Pro chip.
AI-written from the coverage · updated 16 d ago
Coverage timeline
Follow the coverage to see the story from different sides.
- arXiv Game AIPickPlayGen2: Learning real-time playable worlds with explicit 3D state
The authors propose PlayGen2, a real-time interactive world model that separates learned 3D state evolution from neural rendering, using a volumetric occupancy grid and an identity-indexed 3D state to maintain persistent world memory. Tested on Doom, the model learned world dynamics including geometry-aware player movement and player-NPC interaction, and the full simulation and rendering pipeline runs at about 160ms backend latency on a MacBook Pro with an M4 Pro chip.