Genesis
Genesis Embodied AIGenesis is a simulation platform for robotics and embodied AI that combines a multi-physics engine (rigid bodies, cloth, fluids, coupled solvers), its own photorealistic renderer Nyx alongside a ray tracer, and the Quadrants compiler, which lowers Python kernels to CUDA, ROCm, Metal, Vulkan and CPU. It runs many heterogeneous environments in parallel and is developed by Genesis-Embodied-AI.
The repository was renamed from Genesis to genesis-world; the old name redirects to it.
Openness
5 high confidence- license
- Apache-2.0(LICENSE body)
- source
- public
- core features withheld
- no
Apache-2.0 throughout, with the renderer and compiler it builds on published in their own open repositories and no paid edition.
- https://raw.githubusercontent.com/Genesis-Embodied-AI/genesis-world/main/LICENSE recorded 2026-09-27
LICENSE: the Apache License, Version 2.0 text, with no exception clause.
- https://raw.githubusercontent.com/Genesis-Embodied-AI/genesis-world/main/README.md recorded 2026-09-27
README: no pricing, subscription or hosted-service language; the renderer and compiler are separate open repositories it links to.
- https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Genesis-Embodied-AI%2Fgenesis-world recorded 2026-09-27
Repository record for Genesis-Embodied-AI/genesis-world: license apache-2.0, archived false, fork false; previous name genesis-embodied-ai/genesis.
- https://ungh.cc/repos/Genesis-Embodied-AI/genesis-world/files/main recorded 2026-09-27
Full listing of the main branch, 1,058 paths; no enterprise, ee or commercial path.
Adoption
2 medium confidencePyPI installs of genesis-world. PyPI carries no repository link for the package, but its description links issues and images to this repository, so the package is the product.
- https://pypi.org/pypi/genesis-world/json recorded 2026-09-27
PyPI metadata for genesis-world: project_urls empty; the long description links https://github.com/Genesis-Embodied-AI/genesis-world/issues.
- https://pypistats.org/api/packages/genesis-world/recent recorded 2026-09-27
pypistats: last_month 99,572 downloads for genesis-world.
Capability
4 medium confidenceGenesis matches Isaac Lab on parallel simulation, physics breadth and rendered sensors, and goes further on hardware, running on AMD, Apple and Vulkan GPUs as well as NVIDIA. Its README offers demos rather than a library of ready-to-train tasks for common robots, and names no route for checking a trained policy before it goes to hardware.
- https://raw.githubusercontent.com/Genesis-Embodied-AI/genesis-world/main/README.md recorded 2026-09-27
README: "combines a unified multi-physics engine, a photo-realistic renderer (Nyx), and a cross-platform compiler (Quadrants)"; "parallel and heterogeneous environments"; "Quadrants lowers Python kernel code to CUDA, AMD ROCm, Apple Metal, Vulkan, x86, and ARM64"; sensor demos for depth camera, IMU, lidar, tactile and contact force.
Verified 2026-09-27