by World Labs
World Labs' world model that generates and reconstructs explorable 3D worlds from images, text, video or 3D input.
Updated EU status checked Last shown
Sources: the vendor's own pages (linked above) and creator videos at the linked timestamps. EU status is information, not legal advice; see how we check.
Atlas is a world model from World Labs, announced on 12 November 2026 as its next-generation model. It turns images (photos or generated) into 3D worlds that can be explored and edited, including generating new viewpoints and rendering along user-designed camera paths. According to YouTube coverage, it accepts text, image, video or 3D input and can produce up to one minute of coherent 1440p video along a camera path, with outputs such as point clouds and Gaussian splats.
World Labs presents Atlas through several tools: Viewfinder (scene reconstruction and camera paths), Chisel (controllable scene building and grayboxing), Splats to Mesh (conversion of splats into meshes for games), and Relight (lighting and manipulation of splats). A developer API and integrations, including Blender, let users bring Atlas capabilities into their own applications and workflows. A real-time demo, Frameboy, was published on 5 November 2026.
Access appears to be staged: creators described it as a preview or early access, with an access form. The provided pages do not state pricing, weights, or a license, and no open-source release is indicated.
Watch the moment: the exact second top creators show it. Videos load only when you press play, via youtube-nocookie.com.
World model taking text, images, video or 3D input to create or reconstruct 3D worlds, with up to one minute of coherent 1440p video along user-drawn camera paths and outputs like point clouds and Gaussian splats. Currently a preview with early access form.
World Labs model that takes one or more images plus a camera path and reconstructs explorable 3D scenes with pixel-perfect camera control. In early access, and the creator was impressed.
Boston Dynamics showed a robot picking up and carrying a whole fridge; the fridge showed some damage.
Closest neighbours by meaning, not just by category.
Library · World Labs
World Labs tooling that turns a single image into an editable 3D environment with meshes, splats, physics and lighting.
Model
World model that turns a few context images into an editable, physics-simulated 3D world rendered by a video model.
Model · Tencent ARC Lab
Video world model from Tencent ARC Lab that turns an image or text into camera-controlled explorable scenes with 3D-aware memory.
Model
Research framework that turns posed images with instance masks and 3D boxes into per-object meshes for cluttered scenes.
Model
Research model that predicts layered, pixel-aligned 3D point clouds, including hidden geometry, from one image or video.
Model
Research model that unifies 3D scene generation and reconstruction in a single pixel-space diffusion model.