Genesis: A Generative and Universal Physics Engine for Robotics and Beyond
Genesis is an open-source physics simulation platform with massively-parallel GPU acceleration and provisions for generative automation .
As per the project's user guide and GitHub repo , it is designed for robotics & embodied AI by a broad consortium of organizations:
- A universal physics engine re-built from the ground up, capable of simulating a wide range of materials and physical phenomena.
- A lightweight, ultra-fast, pythonic, and user-friendly robotics simulation platform.
- A powerful and fast photo-realistic rendering system.
- A generative data engine that transforms user-prompted natural language description into various modalities of data.
Special thanks to Johnny NΓΊΓ±ez Cano for porting Genesis to arm64+CUDA and Jetson, along with the extensive stack to support it.
What you need
-
One of the following Jetson devices:
Jetson AGX Orin (64GB) Jetson AGX Orin (32GB)
-
Running one of the following versions of JetPack :
JetPack 6 (L4T r36.x)
-
NVMe SSD highly recommended for storage speed and space
-
11.1GB
fordustynv/genesis:r36.4.0-cu128
container image -
Space for models and datasets (
>50GB
)
-
-
Clone and setup
jetson-containers
:git clone https://github.com/dusty-nv/jetson-containers bash jetson-containers/install.sh
-
Please refer to the upstream project for more resources:
- Code Examples - https://github.com/Genesis-Embodied-AI/Genesis/tree/main/examples
- Getting Started - https://genesis-world.readthedocs.io/en/latest/user_guide/index.html
- Documentation - https://genesis-world.readthedocs.io
Work in Progress
Development status of the arm64 porting efforts (2/17/25)
- OMPL - support for motion planning and IK, coming soon to Jetson.
- LuisaRender - support for CUDA-accelerated raytracing, coming soon to Jetson.
-
Ask questions in
#robot-simulation-discuss
on Discord orjetson-containers/issues
Start Container
Currently a container image is available for JetPack 6.1+ and L4T R36.4, built with CUDA 12.8 and PyTorch 2.6:
jetson-containers run dustynv/genesis:r36.4.0-cu128
To mount your own directories into the container, use the
-v
or
--volume
flags:
jetson-containers run -v /path/on/host:/path/in/container dustynv/genesis:r36.4.0-cu128
Demo Walkthroughs
There are numerous
quickstart guides
and
code examples
that come with Genesis and are installed inside the container under
/opt/genesis
jetson-containers run -w /opt/genesis/examples \
dustynv/genesis:r36.4.0-cu128 \
python3 elastic_dragon.py --vis
We encourage you to explore these resources and get involved in the efforts, and share results or feedback with the group and upstream maintainers. We'll continue updating this page with progress from porting the remaining dependencies, along with integration of AI & robotics models and related SDKs like ROS and LeRobot .