Genesis World is a physics simulation platform built for general-purpose robotics and embodied AI learning. It combines a high-speed parallel physics engine with a Pythonic API, so researchers can generate the large-scale training data and interactive environments that robot learning demands, all on accessible hardware.
Genesis World has attracted more than 30,000 stars on GitHub and is under active development, with updates in October 2026. Training robots in the real world is slow and expensive, so embodied AI depends on simulation that is both physically faithful and fast enough to run thousands of environments at once. Genesis was designed around that need: its solver supports rigid bodies, soft materials, fluids and cloth within one unified framework, and it parallelizes massively on GPUs. On top of the engine sits a generative data pipeline direction, where natural-language descriptions can be turned into simulation assets and tasks, lowering the effort of building new robot training scenarios.
Genesis World is free and open source under the Apache-2.0 license. There is no paid tier; you run it on your own hardware or cloud GPUs.
Genesis World is for robotics researchers, embodied AI labs and simulation engineers who need fast, realistic environments to train and evaluate robot policies without the cost and risk of physical hardware trials.