Gymnasium-Robotics
- Software
- Humanoids and robot learning
- Library
A Python library of reinforcement learning robot environments built on the Gymnasium API and the MuJoCo physics engine, covering manipulation, dexterous hands, multi-agent and maze tasks.
Made by Farama Foundation.
Source on GitHub: github.com/Farama-Foundation/Gymnasium-Robotics (opens in a new tab). Facts read at commit b8da02a (opens in a new tab).
Licenses
- Code
- MIT (opens in a new tab)
Read by Robotheca from LICENSE (opens in a new tab) (MIT) at commit b8da02a (opens in a new tab), on 16 Oct 2026.
- Commercial use
- Yes
What it does
- Provides Fetch arm manipulation tasks such as reach, push, slide and pick and place.
- Offers Shadow Dexterous Hand environments for object manipulation with cube, egg-object or pen, some with touch sensors.
- Includes MaMuJoCo multi-agent factorizations of Gymnasium MuJoCo environments using the PettingZoo ParallelEnv API.
- Adds refactored D4RL environments: maze navigation, Adroit Arm tasks and the Franka Kitchen multitask setting.
- Uses a multi-goal API (GoalEnv) with dictionary observations and reward, terminated and truncated functions that can be recomputed with substituted goals, useful for Hindsight Experience Replay.
- Installs with pip; the older mujoco-py based versions are available through an optional extra.
Written by Robotheca from the project's README (opens in a new tab) at commit b8da02a.
Getting started
- Difficulty
- Intermediate
- Hardware it needs, as the project states it
- It states no GPU or memory requirements.
- How to start
- The project's website (opens in a new tab)
Maintenance
- Status
- Active: a commit in the last 90 days
- Newest commit
- 9 Oct 2026 on main
- Latest release
- 1.4.2 (2 Jan 2026)
Checked on GitHub on 16 Oct 2026.
Links
Robots on Robotheca that use it
No robot on Robotheca is linked to it yet. Links come from a robot's robot.yaml, from its dependency files read as text, or from Robotheca's editors.
Robotheca lists software as a reference: nothing is hosted or run here, and every fact links to where it was read.