Atlas: A World Model for Spatial Intelligence

World Labs introduced Atlas, a new omni world model that natively operates on text, images, video, and 3D data to generate, reconstruct, and simulate worlds with spatial intelligence. Atlas is a multimodal autoregressive diffusion transformer that can perform tasks including camera-controlled video generation (up to 1 minute at 1440p), spatial reconstruction of real-world scenes, space-time simulation, and image generation from text prompts. The model scales with increased training compute and will power future versions of World Labs' products, with early access now available.