CAMEL

Active
GitHub Python Apache-2.0

Description

CAMEL is an open-source framework for multi-agent collaboration, supporting role-play, task decomposition, and coordinated execution.

Key Features

  • Scalable multi-agent systems supporting up to 1 million agents
  • Role-playing and task decomposition with coordinated agent execution
  • Stateful memory enabling multi-step interactions and historical context
  • Data generation pipeline: CoT, Self-Instruct, Source2Synth, Self-Improving CoT
  • Built-in tools: RAG pipeline, web browsing, code execution, and knowledge graphs
  • Role-playing society simulation for studying emergent agent behaviors

Use Cases

💡 Researching scaling laws and emergent behaviors in large-scale multi-agent systems
💡 Generating synthetic training data using CoT and Self-Instruct pipelines
💡 Building role-playing simulations for social science and economic modeling
💡 Automating complex task decomposition across specialized agent teams
💡 Developing world simulation environments for agent interaction studies

Strengths & Limitations

Strengths

  • Actively maintained, recent updates
  • High community interest (17.7k stars)
  • Permissive open-source license (Apache-2.0)
  • Established track record (3 years in production)

Quick Start

pip install camel-ai

from camel.agents import ChatAgent
from camel.messages import BaseMessage

agent = ChatAgent(system_message='You are a helpful assistant.')
response = agent.step(BaseMessage(role_name='user', content='Hello!'))
print(response.msgs[0].content)

Related Projects

MetaGPT

70.2k · Python
Stale B

The Multi-Agent Framework for building the first AI Software Company, enabling natural language programming with multi-role collaboration for automated requirement analysis, design, coding, and testing.

multi-agentframeworkpython +2
  • · Multi-role collaboration - Built-in product manager, architect, engineer roles simulating a full software company SOP pipeline
  • · Natural language programming - Input a one-line requirement to auto-generate user stories, competitive analysis, API design, and complete code
  • · Data Interpreter - Dynamic code generation and data analysis for CSV processing, visualization, and exploratory tasks

Ouroboros

5.8k · Python
Active A

Ouroboros is a spec-driven multi-agent framework that shifts from traditional prompting to specification-driven development, supporting multi-agent collaboration, MCP tool integration, and automated workflow orchestration for building high-quality agent systems.

multi-agentframeworkagent +4
  • · Spec-first workflow: Socratic interview → immutable seed spec → execution → evaluation
  • · Multi-runtime adapter supporting Claude Code, Codex CLI, OpenCode, Hermes, Gemini, Kiro, Copilot
  • · Three-layer architecture: Kernel (contract engine), Plugins (domain workflows), Shell (TUI cockpit)

DeepResearchAgent

3.5k · Python
Stale B

DeepResearchAgent is a hierarchical multi-agent system designed for deep research tasks and general-purpose problem solving, using a top-level planning agent to coordinate specialized sub-agents for automated task decomposition and efficient cross-domain execution.

multi-agentframeworkpython +3
  • · Hierarchical multi-agent system with top-level planning agent coordinating specialized sub-agents
  • · RSPL protocol layer modeling prompts, agents, tools, environments, and memory as versioned resources
  • · SEPL self-evolution protocol for proposing, assessing, and committing improvements with rollback support

AG2

4.9k · Python
Active A+

AG2 (formerly AutoGen) is an open-source AgentOS providing a multi-agent conversation framework with flexible agent orchestration, tool integration, and distributed collaboration for building complex multi-agent systems.

multi-agentframeworkagent +3
  • · Multi-agent conversation framework with flexible orchestration patterns including swarms, group chats, and nested chats
  • · Support for multiple LLM providers with configurable API keys and model selection
  • · Human-in-the-loop workflows enabling human oversight and input in agent conversations