mjlab: A Lightweight Framework for GPU-Accelerated Robot Learning

TL;DR

mjlab is a lightweight, open-source framework that enables efficient GPU-accelerated robot learning through modular environments and minimal setup, facilitating rapid development of robotic tasks.

Contribution

It introduces a novel, easy-to-install framework combining GPU simulation, modular API, and native MuJoCo access, simplifying robot learning workflows.

Findings

Supports GPU-accelerated physics simulation

Provides reference implementations for common robotic tasks

Requires minimal dependencies and setup

Abstract

We present mjlab, a lightweight, open-source framework for robot learning that combines GPU-accelerated simulation with composable environments and minimal setup friction. mjlab adopts the manager-based API introduced by Isaac Lab, where users compose modular building blocks for observations, rewards, and events, and pairs it with MuJoCo Warp for GPU-accelerated physics. The result is a framework installable with a single command, requiring minimal dependencies, and providing direct access to native MuJoCo data structures. mjlab ships with reference implementations of velocity tracking, motion imitation, and manipulation tasks.