Vuer: An Event-Driven, Declarative Visualization Toolkit for GenAI and Robotics

pip install 'vuer[all]'   ⬝   docs   ⬝   development

Vuer is a light-weight visualization toolkit for interacting with dynamic 3D and robotics data. It is VR and AR ready, and can be run on mobile devices.

Latest Updates

• 2025-11-28: All documentation examples are now executable with downloadable assets.

Installation

You can install vuer with pip:

pip install -U 'vuer[all]'

Here is an example that loads a URDF file and displays it in the browser. For more examples, see the the examples page.

from vuer import Vuer
from vuer.schemas import DefaultScene, Urdf, OrbitControls

app = Vuer()

@app.spawn(start=True)
async def main(sess):
    sess.set @ DefaultScene(
        Urdf(src="https://raw.githubusercontent.com/nasa-jpl/m2020-urdf-models/main/rover/m2020.urdf"),
        up=[0, 0, -1],  # Z-down coordinate system
        bgChildren=[OrbitControls(key="OrbitControls")]
    )

    await sess.forever()

To get a quick overview of what you can do with vuer, check out the following:

• browse the example gallery here
• try the demo showing a Unitree Go1 robot in front of a staircase here

For more details:

• A full list of visualization components: API documentation on Components.

• A full list of data types: API documentation on Data Types.

Examples

To run the examples, you'll need to download the required assets:

1. Download vuer_doc_assets from this Google Drive link
2. Unzip the downloaded file
3. Place the vuer_doc_assets folder alongside the project directory and rename it to assets.

parent_directory/
├── vuer/                 # This project
│   ├── docs/
│   ├── vuer/
│   └── README.md
└── assets/      # Downloaded assets folder

4. Run the examples:

cd docs/examples/meshes
python mesh_loading.py

Development

Setup

Using uv (recommended):

uv sync --group dev
source .venv/bin/activate

Using pip:

pip install -e '.[dev]'

Common Tasks

make docs     # Build documentation
make preview  # Build and live preview at http://localhost:8000/
make test     # Run tests
make clean    # Clean build artifacts

Contributing

We welcome contributions! See CONTRIBUTING.md for detailed guidelines on:

• Setting up your development environment
• Code quality standards (ruff formatting and linting)
• Documentation workflow
• Publishing releases

To Cite

@software{vuer,
  author = {Ge Yang},
  title = {{VUER}: An Event-Driven, Declarative Visualization Toolkit for GenAI and Robotics},
  version = {},
  publisher = {GitHub},
  url = {https://github.com/vuer-ai/vuer},
  year = {2025}
}

About Us

Vuer is built by researchers at MIT and UCSD in fields including robotics, computer vision, and computer graphics.