# google-deepmind/pysc2

StarCraft II Learning Environment

Repository: https://github.com/google-deepmind/pysc2
Canonical: https://ross.abutalabs.com/products/pysc2
Language: Python
License: Apache-2.0
License Family: permissive
Topics: reinforcement-learning, machine-learning, starcraft-ii, starcraft-ii-replays, deepmind, blizzard-api
Last push: 2024-07-23T16:54:42+00:00

## Health v2 (maintenance only)
Score: 23/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 8, longevity 100
- inputs: {"age_days": 3326, "days_push": 771, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 8307, forks 1163 (observed 2026-08-28T04:10:20.245854+00:00)

## What it is
PySC2 is DeepMind's Python component of the StarCraft II Learning Environment, exposing Blizzard's StarCraft II Machine Learning API as a Python reinforcement learning environment. It lets RL agents observe game state and send actions in StarCraft II, including mini-game maps for research.

## Use cases
- train reinforcement learning agents in StarCraft II
- build an RL environment around a real-time strategy game
- run StarCraft II mini-games for AI research
- parse StarCraft II replays for machine learning
- benchmark deep RL algorithms on a complex game environment

## When to choose
- you need a rich, well-established RL environment with complex multi-agent dynamics
- you want to reproduce DeepMind's StarCraft II RL research
- you need programmatic observation/action interfaces to StarCraft II via Blizzard's API

## When to avoid
- you need a lightweight or fast-to-iterate RL environment
- you don't want to install the full StarCraft II game binary
- you need a project with frequent updates and active development

## Facets
- artifact type: library
- maturity: maintenance
- function: reinforcement-learning, machine-learning, simulation, sdk
- domain: reinforcement-learning, machine-learning, artificial-intelligence
- platform: python, windows
- tags: starcraft-ii, rl-environment, gym-environment, blizzard-api, deepmind, game-ai, research, game-development, linux, macos

## Member repositories
- google-deepmind/pysc2 (main) score 23

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:10:20.245854+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 2026-08-29T17:28:04.140231+00:00, confidence not recorded.
  - readme: https://github.com/google-deepmind/pysc2 (fetched 2026-08-28T04:10:20.245854+00:00, sha 3d981029c568)
- Data as of 2026-08-30T08:39:29.467469+00:00.