OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Multi-Agent Systems

Chatarena

Chatarena is a Python library of multi-agent language-game environments for researching, benchmarking and training LLM agents. It provides game and player abstractions, a framework for interaction loops, and Web UI and CLI interfaces for agent development.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Chatarena — OpenViST project profile in Multi-Agent Systems.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Chatarena when a research or development task needs a Python framework for simulated interactions among multiple LLM agents, especially to explore communication strategies or experiment with game environments. Consider its deprecated status when deciding whether it suits work that depends on ongoing updates or support.

Who it is for

  • Researchers studying social interaction and communication among autonomous LLM agents.
  • Developers who want to prompt, develop or benchmark agents in simulated language games.
  • Teams exploring multi-agent reinforcement-learning experiments using language-game environments.

Documented uses

  • Run a language game in which several LLM agents observe and respond to game state through natural-language interactions.
  • Use the Web UI or CLI to develop and prompt-engineer agents.
  • Define new game dynamics by creating an environment, or customize the Arena to alter the interaction loop.
  • Explore language-game environments for understanding, benchmarking or training agent LLMs.
Before you choose

Requirements & limits

The repository identifies the project as open source under the Apache-2.0 license. The README documents installation through pip or from source.

  • The repository states that ChatArena was deprecated on August 11, 2025, and will receive no further planned updates or support.
  • The README documents Python 3.7 or newer as an installation requirement.
  • Core installation includes only dependencies needed for core functionality; backend and environment integrations are provided through optional dependency groups.
  • An OpenAI API key is optional for installation but required for the README's GPT-3.5-turbo or GPT-4 agent examples.
  • The project offers both Web UI and CLI interfaces, while its Python API supports loading a game from a configuration file.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Install ChatArena and select optional components

Set up the Python package with core dependencies or add documented backend and environment dependencies.

Before you start

  • Python 3.7 or newer.
  • An OpenAI API key is optional for using the README's GPT-3.5-turbo or GPT-4 agent examples.

Steps

  1. Install the package with the command below.

    Documented command
    pip install chatarena

  2. For integrations with all supported language backends, install chatarena[all_backends]; for dependencies for all environments, install chatarena[all_envs].

  3. To install all documented optional dependencies, use the command below.

    Documented command
    pip install chatarena[all]

Expected outcome

The package is installed with core dependencies, or with the selected optional backend and environment dependencies.

Guide limitations

  • The project was deprecated on August 11, 2025, and is no longer planned to receive updates or support.
  • An OpenAI API key is required for the README's GPT-3.5-turbo or GPT-4 agent examples, but is not listed as a general installation requirement.
Documentation 1 ↗
Source-based practical guide

Configure an Arena game run from a repository example

Use the documented Python API example to load a provided game configuration and start its interaction loop.

Before you start

  • ChatArena installed in a Python environment.
  • A local copy of the repository containing examples/nlp-classroom-3players.json.
  • An OpenAI API key if the configured players use the README's GPT-3.5-turbo or GPT-4 agent examples.

Steps

  1. Load the example configuration with Arena.from_config("examples/nlp-classroom-3players.json").

  2. Call arena.run(num_steps=10) to run the example for ten steps.

  3. For interactive command-line use, call arena.launch_cli().

Expected outcome

The configured Arena runs for ten steps, or opens its interactive CLI.

Guide limitations

  • The README presents this as a code example; it does not specify the example's exact game behavior or results.
  • The project is deprecated and no longer planned to receive updates or support.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
13 October 2026
Original website title
Farama-Foundation/ChatArena · Official homepage title · retained in its source language
How we review sourcesSuggest a correction