BigHugger
GH Repository · Farama-Foundation

ChatArena

ChatArena (or Chat Arena) is a Multi-Agent Language Game Environments for LLMs. The goal is to develop communication and collaboration capabilities of AIs.

stars
1,562
30-day movement
starts with the next reading
Related entries
60
Connections
2
Pythongpt-4chatgptpythonaimulti-agent-simulationagent-frameworkartificial-intelligencemulti-agent-reinforcement-learningnatural-language-processingmulti-agentlarge-language-models

ChatArena is a Python framework providing multi-agent language game environments for large language models. Its stated goal is to develop AI communication and collaboration capabilities through game-based interaction.

Use it when you want to run multiple LLM agents against each other in language games to study or train communication and collaboration behavior.

Use it to

  • Set up multi-agent language game environments
  • Evaluate LLM communication abilities
  • Run multi-agent reinforcement learning experiments
  • Simulate agent collaboration scenarios
  • Test GPT-4/ChatGPT agents in interaction

For Researchers in multi-agent LLM and reinforcement learning

Role
agent-framework
Language
Python
Licence
Apache-2.0
Forks
151
Open issues
5
Last push
2025-08-11
Latest release
v0.1.8 · 2023-04-13
topicslarge-language-modelsmulti-agentreinforcement-learningsimulationpythonnlp