# Farama-Foundation/PettingZoo

A standard API for multi-agent reinforcement learning environments, with popular reference environments and related utilities

Repository: https://github.com/Farama-Foundation/PettingZoo
Canonical: https://ross.abutalabs.com/products/pettingzoo
Homepage: https://pettingzoo.farama.org
Language: Python
License: MIT
License Family: permissive
Topics: api, gymnasium, multi-agent-reinforcement-learning, reinforcement-learning, multiagent-reinforcement-learning
Last push: 2026-08-13T18:03:26+00:00

## Health v2 (maintenance only)
Score: 89/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 97, release rhythm 73, longevity 100
- inputs: {"age_days": 2417, "days_push": 20, "days_rel": 20, "gap_med": 108, "n_releases_24m": 4}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 3498, forks 519 (observed 2026-08-28T04:08:07.384327+00:00)

## What it is
PettingZoo is a Python library providing a standard API for multi-agent reinforcement learning environments, analogous to Gymnasium but for multiple agents. It ships with reference environment families (Atari, Butterfly, Classic, SISL) plus wrappers and utilities for MARL research.

## Use cases
- train multi-agent reinforcement learning agents
- benchmark MARL algorithms on standard environments
- create custom multi-agent RL environments with a standard API
- run multi-player Atari and board game environments for RL research
- integrate MARL environments with CleanRL, Tianshou, RLlib, or Stable-Baselines3

## When to choose
- you need a standardized, well-maintained API for multi-agent RL environments
- you want ready-made reference environments like chess, Hanabi, or cooperative Atari games
- you are doing MARL research and want compatibility with popular RL training libraries

## When to avoid
- you only need single-agent RL environments (use Gymnasium instead)
- you need official Windows support
- you need high-performance production simulation rather than research environments

## Facets
- artifact type: library
- maturity: active
- function: reinforcement-learning, simulation, game-engine, testing
- domain: reinforcement-learning, machine-learning, artificial-intelligence
- platform: python
- tags: multi-agent, gymnasium, rl-environments, marl, farama-foundation, game-development, research, linux, macos

## Member repositories
- Farama-Foundation/PettingZoo (main) score 89

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:07.384327+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 2026-08-29T18:35:50.799644+00:00, confidence not recorded.
  - readme: https://github.com/Farama-Foundation/PettingZoo (fetched 2026-08-28T04:08:07.384327+00:00, sha 285b83a55bc8)
  - homepage: https://pettingzoo.farama.org (fetched 2026-08-29T09:29:55.837722+00:00, sha d2bd582b7b1f)
  - registry_pypi: https://pypi.org/pypi/pettingzoo/json (fetched 2026-08-29T09:29:55.846664+00:00, sha 087b5b5e4ed7)
- Data as of 2026-08-30T08:39:29.467469+00:00.