google-deepmind/dm_control

Google DeepMind's software stack for physics-based simulation and Reinforcement Learning environments, using MuJoCo.

73
/ 100
Verified

This tool provides a powerful set of capabilities for designing and running physics-based simulations, especially for developing and testing AI agents. It takes in descriptions of physical models and environments, and outputs simulated interactions and observations. Researchers and engineers working on robotics, control systems, and artificial intelligence will find this useful for creating realistic virtual testbeds.

4,494 stars. Used by 2 other packages. Actively maintained with 3 commits in the last 30 days. Available on PyPI.

Use this if you need to build complex, physically accurate simulated environments to train and evaluate reinforcement learning agents.

Not ideal if you are looking for a simple, drag-and-drop simulation tool or do not have programming experience.

robotics simulation reinforcement learning control systems AI agent training virtual prototyping
Maintenance 13 / 25
Adoption 12 / 25
Maturity 25 / 25
Community 23 / 25

How are scores calculated?

Stars

4,494

Forks

742

Language

Python

License

Apache-2.0

Last pushed

Mar 11, 2026

Commits (30d)

3

Dependencies

15

Reverse dependents

2

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/google-deepmind/dm_control"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.