---
id: mujoco
version: "3.11.0"
license: Apache-2.0
license_treatment: permissive
maintenance: active
---

# mujoco — MuJoCo Physics Simulator

License: permissive · Maintenance: active · Downloads: 3.1M/mo

## What it is and what it does

MuJoCo is Google DeepMind's physics simulation engine exposed through Python bindings that give direct access to the underlying C API. It provides structs, constants, and enumerations for rigid-body dynamics, collision detection, and constraint solving. The package bundles the MuJoCo library itself, so no separate installation is needed. It depends on numpy, absl-py, etils, glfw, and pyopengl for rendering utilities.

The bindings are designed as a low-level interface rather than a high-level scene authoring API. They include utilities for setting up OpenGL rendering contexts (egl, glfw, osmesa subpackages) and are commonly used in robotics research, reinforcement learning, and physics-based simulation workflows. The package is actively maintained by DeepMind and kept synchronized with MuJoCo's latest developments.

Use it for:

- Build reinforcement learning environments with physics-accurate robot simulation and control
- Prototype rigid-body dynamics for robotics applications with direct C API access
- Render physics simulations using OpenGL with built-in context setup utilities
- Conduct biomechanics or mechanical system research requiring precise collision and constraint solving
- Integrate physics simulation into game engines or interactive applications

## Worth the install?

AI-flagged interpretation of the facts above — verify before relying.

MuJoCo is a physics simulation engine with Python bindings that provides direct access to the C API for rigid-body dynamics, collision detection, and rendering.

Yes, if you need a physics engine for robotics, reinforcement learning, or scientific simulation. The package is actively maintained, has no known vulnerabilities, carries a permissive Apache-2.0 license, and provides prebuilt wheels for common platforms. Medium install friction is typical for compiled bindings. Not necessary if you only need a higher-level scene API—consider wrapper libraries instead.

## Install

pip install mujoco
uv add mujoco
poetry add mujoco

## Installing mujoco

Before you install: Medium install friction due to compiled wheels for multiple architectures and Python versions (3.10–3.14). Active maintenance with a recent release (17 days old) and strong repository activity (14550 stars). No known vulnerabilities.

License in practice: Apache-2.0 permissive license allows commercial and private use with attribution. Pre-built wheels include third-party code (Abseil, cereal, Eigen, Collisions) under compatible licenses (Apache-2.0, BSD, MPL2).

Quickstart:

pip install mujoco
import mujoco
model = mujoco.MjModel.from_xml_path('model.xml')
data = mujoco.MjData(model)
mujoco.mj_step(model, data)

Requires Python 3.10 or later. OpenGL rendering requires a display or headless context setup via the egl, glfw, or osmesa subpackages.

Verify before relying:

- Whether the package includes pre-built MuJoCo binaries or requires separate installation
- Performance characteristics and simulation accuracy compared to other physics engines
- Specific use-case maturity for robotics vs. general rigid-body simulation

## Package facts

- License: Apache-2.0 (permissive)
- Python support: supports_current
- Install friction: medium
- Maintenance: active
- Downloads: 3.1M/month (top 5,000 on PyPI)
- Known vulnerabilities: none known

## Tags

physics simulation engine, rigid body dynamics, collision detection python, physics engine bindings, robotics simulation, mujoco physics, dynamics simulation library, physics-simulation, robotics, reinforcement-learning

[View on SkillFed](https://skillfed.io/packages/mujoco) · [View on PyPI](https://pypi.org/project/mujoco/)