LocoMuJoCo
- Software
- Humanoids and robot learning
- Library
An imitation learning benchmark for whole-body control, with humanoid, quadruped and musculoskeletal human environments built on MuJoCo and JAX, plus datasets of motion capture data.
Source on GitHub: github.com/robfiras/loco-mujoco (opens in a new tab). Facts read at commit 3921fed (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 3921fed (opens in a new tab), on 16 Oct 2026.
- Commercial use
- Yes
What it does
- Provides a single MuJoCo environment and parallel MJX/MJWarp environments for training.
- Includes 12 humanoid and 4 quadruped environments, four of them biomechanical human models.
- Offers clean single-file JAX implementations of PPO, GAIL, AMP and DeepMimic for benchmarking.
- Ships over 22,000 motion capture datasets (AMASS, LAFAN1 and native) retargeted to each humanoid.
- Supports retargeting datasets from one robot to another and comparing trajectories with DTW and discrete Fréchet distance.
- Exposes a Gymnasium interface, built-in domain and terrain randomization, and modular reward and observation components.
Written by Robotheca from the project's README (opens in a new tab) at commit 3921fed.
Getting started
- Difficulty
- Intermediate
- How to start
- The project's website (opens in a new tab)
Maintenance
- Status
- Maintained: a commit in the last 12 months
- Newest commit
- 10 Mar 2026 on master
- Latest release
- LocoMuJoCo v1.1.0 (10 Mar 2026)
Checked on GitHub on 16 Oct 2026.
Links
Robots that can run 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.