---
id: drake
version: "1.55.0"
license: Various
license_treatment: permissive
maintenance: active
---

# drake — Model-based design and verification for robotics

License: permissive · Maintenance: active · Downloads: 265.4K/mo

## What it is and what it does

Drake is a robotics toolbox developed by the MIT Computer Science and Artificial Intelligence Lab and now led by the Toyota Research Institute. It provides tools for analyzing robot dynamics and designing control systems with an emphasis on optimization-based approaches. The package integrates with standard scientific Python libraries (numpy, matplotlib) and includes visualization via pydot and configuration via PyYAML.

It is primarily used by roboticists and control engineers to model, simulate, and verify robotic systems. The package targets modern Python (3.12+) and is distributed as compiled wheels for Linux and macOS, making installation straightforward on supported platforms. Its heavy reliance on optimization—particularly via the Mosek solver—means it is best suited for users working on trajectory planning, motion control, and system verification rather than general-purpose robotics tasks.

Use it for:

- Simulate and analyze robot arm dynamics for trajectory planning and control design.
- Verify robot locomotion controllers using optimization-based methods.
- Design and test manipulation tasks with dynamics constraints.
- Build and validate multi-robot coordination strategies.
- Prototype and validate control algorithms before hardware deployment.

## Worth the install?

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

Drake is a robotics toolbox for analyzing robot dynamics and building control systems using optimization-based design and verification methods.

Yes, if you are working on robot dynamics, control design, or trajectory optimization and can work with Python ≥3.12. The active maintenance, recent release, and zero known vulnerabilities are positive signals. However, verify that the Mosek solver dependency and mixed licensing (BSD + proprietary) align with your project's requirements and budget before committing.

## Install

pip install drake
uv add drake
poetry add drake

## Installing drake

Before you install: Medium install friction due to compiled wheels for specific Python versions (3.12–3.14) and platforms (Linux, macOS ARM64/x86_64). Active maintenance with a release 30 days ago. Requires Python ≥3.12 and has five runtime dependencies including the commercial solver Mosek.

License in practice: Licensed under BSD and other proprietary terms ("Various"). The permissive treatment suggests no viral copyleft, but the mixed licensing warrants review of the full license terms before commercial use.

Quickstart:

pip install drake
import drake
plant = drake.systems.LeafSystem()

Requires Python ≥3.12; Mosek solver (a runtime dependency) may require a separate license or installation.

Verify before relying:

- Whether Mosek is freely available or requires a commercial license for production use.
- Full scope of "Various" license terms and any restrictions on derivative works or commercial deployment.
- Availability of pre-built wheels for all target platforms and Python versions in your environment.

## Package facts

- License: Various (permissive)
- Python support: supports_current
- Install friction: medium
- Maintenance: active
- Downloads: 265.4K/month (top 15,000 on PyPI)
- Known vulnerabilities: none known

## Tags

robot dynamics simulation, robotics control systems, optimization-based robot design, robot trajectory planning, dynamics analysis toolbox, robotic system verification, motion planning robotics, robotics, dynamics-simulation, control-systems

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