Nova3D

website · open studio · paper · datasets · discord · X · issues

A working farm, written by Nova3D from one prompt. Humanoids pick tomatoes, quadrupeds carry the harvest, drones scout. Simulated and filmed in MuJoCo.

Capabilities

1. Executable source: represent assets and environments as programs that can be inspected, edited, and rebuilt.
2. Parametric control: define dimensions, proportions, materials, physical properties, and motion through explicit parameters.
3. Controlled variation: specify which parameters change and which remain fixed across design alternatives or experiments.
4. Structural editing: change components, assemblies, layouts, clearances, and connections through their construction code.
5. Reusable construction: reuse components and generation logic to build related designs, asset families, and environments.
6. Programmable motion: define articulation, animation, interactive controls, and prescribed object trajectories.
7. Geometry and motion checks: inspect dimensions, part relationships, pivots, and collisions, using checks appropriate to the design.
8. Versioned development: preserve source, parameters, build outputs, and validation results as the project evolves.
9. Target-specific exports: compile designs for CAD tools, 3D applications, and robotics simulators through dedicated export paths.

For robotics

1. Controlled scenario generation: specify which geometric, physical, and temporal parameters change, and which remain fixed.
2. Structural variation: change connectivity, obstacle arrangement, clearances, and mechanism configurations, not just textures or initial positions.
3. Reproducible evaluation: preserve the environment source, parameters, simulator configuration, and experiment results.
4. Failure-driven generation: eventually turn a failed rollout into a family of nearby scenarios to investigate the failure boundary.

Datasets

Our releases include Nova3D-Bench (preview), the benchmark used in our paper, and 20 outdoor simulation environments reconstructed from Wanderland captures, with editable scene source, annotations, and native simulator exports.

Explore the datasets

Describe it. Build it. Keep going.

Start with an idea and keep shaping it. Change the geometry, add motion and behavior, and explore new versions.

Build it together.

People and AI agents work in the same world at the same time, each on their own branch. Then the branches merge into one world.

worlds

Robots at work.

Every world below started as a text prompt, was generated as an executable scene program, compiled to MuJoCo, and populated with robots. Doors open on real hinges, produce carries labeled masses, floors carry measured friction. The environment is code, so every physical property is authored, not painted on.

Simulate. Test. Revise.

Run your designs in simulation to check movement, collisions, and interactions. Find what fails, revise the design, and test it again.

physics

Physics, defined in source.

Mass, inertia, friction, and joint limits are explicit and editable alongside the geometry. Change the conditions and simulate the result. Below, the same scene runs with one parameter changed: gravity, from 0 to 9.81 m/s².

Build on common ground.

Bring assets, scenes, and behavior together in one programmable project. Build from the same source for design tools, game engines, and simulators.

License

MIT © RareSense.