ProjectFeatured
ManiSkill
Large-scale manipulation benchmark and training env with GPU parallelism and diverse tasks.
PythonIntermediateUpdated 2026-08-04
ProjectIntermediate
robosuite
Modular MuJoCo-based robot simulation framework for composing manipulation tasks and algorithms.
Overview
robosuite is a modular MuJoCo-based framework for composing robot manipulation environments. You select robots, grippers, objects, and arenas, then attach learning algorithms. It underpins robomimic and many academic benchmarks.
Choose robosuite when you want composable tabletop manipulation with clean APIs rather than a single fixed benchmark suite.
Who it is for
Academic labs building custom MuJoCo manipulation tasks.
Key highlights
- Modular robot / gripper / object composition
- MuJoCo backend with research adoption
- Foundation for robomimic datasets
When to use
- Custom tabletop sim tasks
- IL/RL papers needing flexible envs
When not to use
- Photoreal or mobile navigation focus
Getting started
- 1Install robosuite and run a Lift or Stack demo.
- 2Compose a custom env from existing components.
- 3Connect BC or RL via Gymnasium wrappers.
How it compares
★ 3.5k
ProjectFeatured
Isaac Lab
NVIDIA robot learning framework on Isaac Sim for large-scale parallel training and sim-to-real.
Latest stable v3.0.0 ↗Caveats & pitfalls
- Visual fidelity is research-grade, not product digital twin.
Content reviewed 2026-07-23
Comments
Share an update or discuss robosuite.
Comments are public; please do not include private information.