openai/multiagent-particle-envs

• highreadme#1

  Reposition README's core value proposition before archive status

  Why:

  CURRENT

  **Status:** Archive (code is provided as-is, no updates expected)
  
  # Maintained Fork
  
  The maintained version of these environments...

  COPY-PASTE FIX

  Copy

  # Multi-Agent Particle Environment (MPE)
  
  This repository provides the original code for the Multi-Agent Particle Environment (MPE), a simple multi-agent particle world with continuous observation and discrete action spaces, along with basic simulated physics. It was used in the seminal paper "Multi-Agent Actor-Critic for Mixed Cooperative-Competitive Environments."
  
  **Status:** This repository is archived and provided as-is, with no further updates expected. For an actively maintained version with numerous fixes, comprehensive documentation, pip installation, and support for current Python versions, please refer to the PettingZoo project (https://github.com/Farama-Foundation/PettingZoo, https://pettingzoo.farama.org/environments/mpe/).

  OpenDoneDismiss
• hightopics#2

  Add specific multi-agent reinforcement learning topics

  Why:

  CURRENT

  paper

  COPY-PASTE FIX

  Copy

  multi-agent-reinforcement-learning, marl, reinforcement-learning, multi-agent-systems, simulation, environments, openai-gym, particle-environments

  OpenDoneDismiss
• mediumabout#3

  Refine the 'About' description for clarity on purpose

  Why:

  CURRENT

  Code for a multi-agent particle environment used in the paper "Multi-Agent Actor-Critic for Mixed Cooperative-Competitive Environments"

  COPY-PASTE FIX

  Copy

  The original multi-agent particle environment (MPE) code, serving as a minimalist testbed for multi-agent reinforcement learning research, particularly for replicating results from the paper "Multi-Agent Actor-Critic for Mixed Cooperative-Competitive Environments."

  OpenDoneDismiss

Category GEO backends resolved for this scan: google/gemini-2.5-flash, deepseek/deepseek-v4-flash

• CATEGORY QUERY

  What are good multi-agent reinforcement learning environments for experimenting with new algorithms?

  you: #3
  AI recommended (in order):

  1. SMAC (StarCraft Multi-Agent Challenge) (oxwhirl/smac)
  2. PettingZoo (Farama-Foundation/PettingZoo)
  3. Multi-Agent Particle Environments (MPE) (openai/multiagent-particle-envs) ← you
  4. Google Research Football (GRF) (google-research/football)
  5. MAgent (PKU-MARL/MAgent)
  6. Neural MMO (neuralmmo/neuralmmo)
  7. OpenSpiel (deepmind/open_spiel)
  Show full AI answer
• CATEGORY QUERY

  Looking for a Python-based multi-agent simulation environment with continuous observations and discrete actions.

  you: not recommended
  AI recommended (in order):

  1. PettingZoo
  2. MAgent
  3. Multi-Agent Particle Environment (MPE)
  4. OpenSpiel
  5. Gymnasium
  6. Pymunk

  AI recommended 6 alternatives but never named openai/multiagent-particle-envs. This is the gap to close.

  Show full AI answer

• Metadata completeness

  pass

• README presence

  pass

• Compared to common alternatives in this category, what is the core differentiator of openai/multiagent-particle-envs?

  pass
  ✓AI named openai/multiagent-particle-envs explicitly

  AI answers can be confidently wrong. Read for accuracy: does it match your actual tech stack, audience, and differentiator?

• If a team adopts openai/multiagent-particle-envs in production, what risks or prerequisites should they evaluate first?

  pass
  ✓AI named openai/multiagent-particle-envs explicitly

  AI answers can be confidently wrong. Read for accuracy: does it match your actual tech stack, audience, and differentiator?

• In one sentence, what problem does the repo openai/multiagent-particle-envs solve, and who is the primary audience?

  pass
  ⚠AI did not name openai/multiagent-particle-envs — likely talking about a different project

  AI answers can be confidently wrong. Read for accuracy: does it match your actual tech stack, audience, and differentiator?