Fei-Fei Li's World Labs unveils Atlas: an omni world model with 1-min 1440p video and 3D reconstruction
theworldlabs · x · 2026-09-09
World Labs introduced Atlas, a next-generation omni world model pretrained from scratch to natively operate on text, images, video, and 3D. It's a multimodal autoregressive diffusion transformer that merges all inputs into a shared spatial context, staying 3D-consistent and scaling with compute.
- Camera-controlled generation: pixel-perfect camera control from one or more images, up to 1 minute of 1440p video
- Spatial reconstruction: from 1 to dozens of images, outperforming specialized 3D reconstruction SOTA with novel-view frames and explicit 3D outputs
- Space-time simulation: video reframing and Real-to-Sim workflows for robotics
- Image generation: text-to-image and 360 panoramas with complex prompts and text rendering
Atlas will power future versions of Marble; early access is now open.
Related event: World Labs unveils Atlas, an omni world model(4 posts)→
More from Embodied
- China now has at least 90 humanoid data collection centers, turning teleop data into its deepest moat — paigeinsf · 2026-09-09
- China Builds 90+ Humanoid Data-Training Centers, Shipped 97% of World's Humanoids — paigeinsf · 2026-09-09
- United Imaging wins CE mark for world's first dual wide-coverage CT family — PAstynome · 2026-09-09
- Fly Escape Room: first game with NPCs powered by a real fruit fly brain, all 166k neurons — juanbenet · 2026-09-09
- MIT TR35 Honoree Shuang Li Uses Language Models and UVA to Make Robots More Adaptable — endernewton · 2026-09-09
- Lightwheel open-sources 100,000 hours of egocentric video for robot training — chris_j_paxton · 2026-09-09