Skip to content

Software

unitree_lerobot

By Unitree Robotics, China (checked on github.com/unitreerobotics (opens in a new tab)), from github.com/unitreerobotics/unitree_lerobot (opens in a new tab)

  • Humanoids and robot learning
  • Framework
  • Apache-2.0

A toolkit for training and testing LeRobot policies on Unitree robots, covering dataset conversion, data editing, model training, and real-world or simulated evaluation.

Facts

GitHub stars
785read on 6 Nov 2026
Last update
25 May 2026Newest commit on main
Maintenance
Maintaineda commit in the last 12 months
Difficulty
Intermediate

Checked on GitHub on 6 Nov 2026.

Licenses

Code
Apache-2.0 (opens in a new tab)

Read by Robotheca from LICENSE (opens in a new tab) at commit 41c2805 (opens in a new tab), on 16 Oct 2026. A license text Robotheca's reader does not know: a person at Robotheca set it on 9 Nov 2026.

Commercial use
Yes

Overview

What it does: Robotheca's summary

  • Loads and visualizes LeRobot datasets, including a published G1 Dex3 dataset from Hugging Face.
  • Provides a PyQt5 data editor for trimming episodes and deleting bad ones.
  • Converts Unitree JSON recordings into the LeRobot dataset format, with several robot type presets.
  • Trains ACT, Diffusion, Pi0, Pi05 and Gr00t policies using the LeRobot training scripts.
  • Runs trained policies on a real G1 robot or in the unitree_sim_isaaclab simulator, with optional visualization.
  • Replays recorded datasets on the robot to validate behavior.

Written by Robotheca from the project's README (opens in a new tab) at commit 41c2805, not by its creator.

The README

Read the README on GitHub (opens in a new tab)

Files and versions

Facts read at commit 41c2805

Robotheca never hosts the code: each link goes to the creator's own page.

Use it

How to install it, in the creator's words: The project's README (opens in a new tab)

Robotheca doesn't review code. Read it before running it.

Robotheca lists software as a reference: nothing is hosted or run here, and every fact links to where it was read.