---
id: robosuite
version: "1.5.2"
license: unclear
license_treatment: unclear
maintenance: active
---

# robosuite — robosuite: A Modular Simulation Framework and Benchmark for Robot Learning

License: unclear · Maintenance: active · Downloads: 252.2K/mo

## What it is and what it does

robosuite is a modular simulation framework built on the MuJoCo physics engine, designed for researchers and practitioners developing robot learning algorithms. It provides a standardized suite of manipulation tasks, procedural environment generation, multiple robot embodiments (including humanoids as of v1.5), and a range of control abstractions—from joint-space velocity control to whole-body controllers. The framework also integrates teleoperation devices, multi-modal sensors (RGB, depth, proprioception), and photorealistic rendering.

The package is actively maintained by researchers at Stanford, UT Austin, and NVIDIA, and is intended to lower barriers to entry for robot learning research by offering reproducible benchmarks, modular APIs for custom environment design, and utilities for collecting and replaying human demonstrations. It supports current Python versions and has no known security vulnerabilities.

Use it for:

- Train reinforcement learning agents on standardized manipulation tasks with reproducible benchmark environments.
- Prototype custom robot environments by composing robot models, arenas, and parameterized objects procedurally.
- Collect and replay human teleoperation demonstrations for imitation learning workflows.
- Evaluate different robot embodiments and controller types on the same task for comparative research.
- Generate synthetic training data with multi-modal sensors (RGB, depth, state) for vision-based robot learning.

## Worth the install?

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

robosuite is a MuJoCo-powered simulation framework for building, benchmarking, and training robot control policies through reinforcement learning and imitation learning.

Yes, if you are doing robot learning research or prototyping. robosuite is actively maintained, has low install friction, zero known vulnerabilities, and is widely used in academic robotics labs. The main caveat is that the license treatment is unclear—verify the license terms before use in proprietary projects. The 12 runtime dependencies are standard and well-maintained.

## Install

pip install robosuite
uv add robosuite
poetry add robosuite

## Installing robosuite

Before you install: Low install friction with a pure-Python wheel and active maintenance (last commit 2026-07-11, 233 days since latest release). Depends on 12 runtime packages including mujoco, numpy, and scipy—all widely available.

License in practice: License treatment is unclear; the fact sheet does not specify an SPDX identifier or raw license text, so you should verify the license terms before use in proprietary or restricted contexts.

Quickstart:

pip install robosuite
import robosuite as suite
env = suite.make(env_name="Lift", robots="Panda", has_renderer=True)
obs = env.reset()
env.render()

Requires mujoco and mink system dependencies; rendering features need a display or headless rendering setup.

Verify before relying:

- Whether the license is open-source or proprietary (license_treatment is marked unclear)
- Whether mujoco and mink dependencies have their own license constraints that cascade to users

## Package facts

- License: not declared (unclear)
- Python support: supports_current
- Install friction: low
- Maintenance: active
- Downloads: 252.2K/month (top 15,000 on PyPI)
- Known vulnerabilities: none known

## Tags

robot simulation framework, reinforcement learning robotics, mujoco robot environments, robot manipulation benchmark, robot learning simulation, imitation learning robotics, procedural robot environments, robotics-simulation, reinforcement-learning, benchmark-environments

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