World model that turns a few context images into an editable, physics-simulated 3D world rendered by a video model.
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.
Mundo is a research world model that builds an interactive 3D environment from 1 to dozens of context images. It reconstructs a 3D proxy of the scene's layout and objects, defines physical properties (mass, friction, pose, size) in a code-based scene program, and uses a physics engine (MuJoCo) to compute motion and interactions. A video model then renders the results, using the supplied images as visual context to preserve appearance.
Described capabilities include rendering new views along user-specified camera paths from sparse views or a single image, long camera paths of tens of seconds to minutes through large scenes via chunked generation with overlapping windows, scripted actions and forces with rendered collisions, and navigation planning (e.g. A*) for robots in Real-to-Sim workflows. Unseen regions are inferred by the model, so more input views reduce invented content.
Based on the project page, no code, weights or paper link are available yet. A YouTube creator reported that code and weights are due at the end of November. Vendor and licence are not stated in the provided material.
Watch the moment: the exact second top creators show it. Videos load only when you press play, via youtube-nocookie.com.
World model that builds an explorable, physics-simulated environment from images, with object properties and physics engine. Code and weights due end of November.
Closest neighbours by meaning, not just by category.
Model · World Labs
World Labs' world model that generates and reconstructs explorable 3D worlds from images, text, video or 3D input.
Model · Runway
Runway's real-time interactive world model generating 720p/24fps video with audio, steered by text actions and camera input.
Library · World Labs
World Labs tooling that turns a single image into an editable 3D environment with meshes, splats, physics and lighting.
Model · Adobe
Adobe Research video world model that builds navigable interactive worlds from an image or video at 16 FPS.
Model
Real-time interactive world model with compressed memory and hybrid action and semantic control, built on Wan 2.2.
Model
Open framework and dataset for turning video diffusion models into real-time, long-horizon interactive world models.
