# google-deepmind/dm_control

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

Repository: https://github.com/google-deepmind/dm_control
Canonical: https://ross.abutalabs.com/products/dm_control
Language: Python
License: Apache-2.0
License Family: permissive
Topics: machine-learning, artificial-intelligence, neural-networks, deep-learning, reinforcement-learning, physics-simulation, mujoco
Last push: 2026-08-20T14:19:29+00:00

## Health v2 (maintenance only)
Score: 94/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 98, release rhythm 86, longevity 100
- inputs: {"age_days": 3169, "days_push": 13, "days_rel": 13, "gap_med": 34, "n_releases_24m": 18}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 4672, forks 761 (observed 2026-08-28T04:08:56.656881+00:00)

## What it is
Google DeepMind's Python stack for physics-based simulation and reinforcement learning environments built on the MuJoCo physics engine. It provides Python bindings to MuJoCo, a benchmark suite of continuous control tasks, an interactive viewer, and libraries for composing custom environments and multi-agent tasks.

## Use cases
- train reinforcement learning agents on continuous control tasks
- simulate physics-based environments with MuJoCo in Python
- benchmark RL algorithms on standard control suites
- build custom RL environments from reusable components
- run multi-agent soccer locomotion tasks
- compose and modify MuJoCo MJCF models programmatically
- visualize physics environments interactively

## When to choose
- you need well-established MuJoCo-based RL benchmark environments
- you want Python bindings and tooling around the MuJoCo physics engine
- you need to compose complex control tasks or multi-agent environments
- you are doing robotics or continuous control research

## When to avoid
- you need GPU-accelerated massively parallel simulation (e.g. for fast RL training at scale)
- you work outside Python
- you need game-engine-quality rendering or non-robotics simulation
- you want editable pip installs of the package

## Facets
- artifact type: library
- maturity: stable
- function: simulation, machine-learning, reinforcement-learning, graphics
- domain: reinforcement-learning, robotics, machine-learning, simulation
- platform: python, windows
- tags: mujoco, physics-engine, rl-environments, continuous-control, deepmind, mjcf, locomotion, multi-agent, linux, macos

## Member repositories
- google-deepmind/dm_control (main) score 94

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:56.656881+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T18:19:22.748944+00:00, confidence not recorded.
  - readme: https://github.com/google-deepmind/dm_control (fetched 2026-08-28T04:08:56.656881+00:00, sha f788390d9b5d)
  - registry_pypi: https://pypi.org/pypi/dm_control/json (fetched 2026-08-29T09:03:35.947504+00:00, sha 469ef2a67200)
- Data as of 2026-08-30T08:39:29.467469+00:00.