robosuite
ARISE Initiativerobosuite is a modular simulation framework and benchmark for robot learning built on the MuJoCo physics engine. It provides manipulation tasks, a range of robot embodiments including humanoids, composite and whole-body controllers, and teleoperation devices for collecting demonstrations. It comes from Stanford's SVL and is maintained with UT Austin and NVIDIA under the ARISE Initiative.
Openness
5 high confidence- license
- MIT(LICENSE body
- source
- public
- core features withheld
- no
MIT throughout, and nothing is held back for a paid edition. GitHub reports the license as 'other' because the file adds an attribution note for the MuJoCo code it uses.
- https://raw.githubusercontent.com/ARISE-Initiative/robosuite/HEAD/LICENSE recorded 2026-09-26
LICENSE: the MIT license text, with an attribution note for the Apache-2.0 MuJoCo code it uses; GitHub labels the file 'other' only because of that note.
- https://raw.githubusercontent.com/ARISE-Initiative/robosuite/HEAD/README.md recorded 2026-09-26
README: no hosted service or paid tier is mentioned.
- https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ARISE-Initiative%2Frobosuite recorded 2026-09-26
Repository record for ARISE-Initiative/robosuite: archived false, fork false.
- https://ungh.cc/repos/ARISE-Initiative/robosuite/files/master recorded 2026-09-26
Full listing of the master branch; no enterprise, ee or commercial path.
Adoption
3 medium confidencePyPI installs of the robosuite package, whose metadata links to this repository. Some of the volume arrives as a dependency of benchmarks built on robosuite.
- https://pypi.org/pypi/robosuite/json recorded 2026-09-26
PyPI metadata for robosuite: Homepage https://github.com/ARISE-Initiative/robosuite.
- https://pypistats.org/api/packages/robosuite/recent recorded 2026-09-26
pypistats: last_month 320,845 downloads for robosuite.
Capability
3 medium confidencerobosuite turns MuJoCo into a place to learn manipulation, adding tasks, controllers, teleoperation for recording demonstrations and photorealistic camera rendering. It simulates one environment at a time on the CPU, so it cannot generate data at the scale the GPU simulators here reach.
- https://raw.githubusercontent.com/ARISE-Initiative/robosuite/HEAD/README.md recorded 2026-09-26
README: "a simulation framework powered by the MuJoCo physics engine for robot learning"; v1.5 "features support for diverse robot embodiments (including humanoids), custom robot composition, composite controllers (including whole body controllers), more teleoperation devices, photo-realistic rendering"; no GPU-parallel claim.
Verified 2026-09-26