AWorld: Advancing Agentic AI

News

• 🦩 [2025/06/19] We have updated our score to 72.43 on the GAIA test. Additionally, we have introduced a new local running mode. See ./README-local.md for detailed instructions.
• 🐳 [2025/05/22] For quick GAIA evaluation, MCP tools, AWorld, and models are now available in a single Docker image. See ./README-docker.md for instructions and youtube video for demo.
• 🥳 [2025/05/13] AWorld has updated its state management for browser use and enhanced the video processing MCP server, achieving a score of 77.58 on GAIA validation (Pass@1 = 61.8) and maintaining its position as the top-ranked open-source framework. Learn more: GAIA leaderboard
• ✨ [2025/04/23] AWorld ranks 3rd on GAIA benchmark (69.7 avg) with impressive Pass@1 = 58.8, 1st among open-source frameworks. Reproduce with python examples/gaia/run.py

Introduction

For self-improving, AWorld (Agent World) is designed to achieve two primary objectives: (1) provide the effiecent forward process, and (2) facilitate diverse backward processes, including but not limit foundation model training and system design meta-learning.

Forward process

1. Agent Construction	2. Topology Orchestration	3. Environments
• ✅ Support different model services • ✅ Support MCP tools • ✅ Support custom tools	• ✅ Encapsulate protocol between models and tools • ✅ Encapsulate protocol among agents	• ✅ Encapsulate runtime state management • ✅ Support state tracing • ✅ Support distributed high-concurrency envs

Follow the instructions in ./README-local.md to run a forward process on the GAIA benchmark. Watch the demo on Youtube

Incubated backward methods

Method Category	Description	Key Information
Foundation Model Training	Improving Function call ability of large language models

Want to build your own multi-agent system? Check out the detailed tutorials below to get started! ⬇️⬇️⬇️

Installation

Python>=3.11:

git clone https://github.com/inclusionAI/AWorld
cd AWorld
python setup.py install

Usage

Here’s a quick start guide to: (1) create your first agent; (2) equip it with a MCP tool; (3) assign a teammate; and (4) answer a user query through teamwork.

from aworld.config.conf import AgentConfig
from aworld.agents.llm_agent import Agent
from aworld.runner import Runners
from aworld.core.agent.swarm import Swarm

if __name__ == '__main__':
    agent_config = AgentConfig(
        llm_provider="openai",
        llm_model_name="gpt-4o",

        # Set via environment variable or direct configuration
        # llm_api_key="YOUR_API_KEY", 
        # llm_base_url="https://api.openai.com/v1"
    )

    # Register the MCP tool here, or create a separate configuration file.
    mcp_config = {
        "mcpServers": {
            "amap-amap-sse": {
                "type": "sse",
                "url": "https://mcp.amap.com/sse?key=YOUR_API_KEY",
                "timeout": 5,
                "sse_read_timeout": 300
            }
        }
    }

    # Create your first agent equipped with an MCP tool
    search = Agent(
        conf=agent_config,
        name="search_agent",
        system_prompt="You are a helpful agent.",
        mcp_servers=["amap-amap-sse"], # MCP server name for agent to use
        mcp_config=mcp_config
    )

    # Add a new teammate to the agent
    summary = Agent(
        conf=agent_config,
        name="summary_agent",
        system_prompt="You are a helpful summary agent."
    )

    # Collaborate as a team; the default is a static workflow
    swarm = Swarm(search, summary)

    # Run agent team
    res = Runners.sync_run(input="Hotels within 1 kilometer of West Lake in Hangzhou",
                     swarm=swarm)
    print(res)

Demo

Running Pre-defined Agents (e.g., see demo code). Below are demonstration videos showcasing AWorld's capabilities across various agent configurations and environments.

Mode	Type	Demo
Single Agent	Browser use	▶️ Watch Browser Demo on YouTube
	Phone use	▶️ Watch Mobile Demo on YouTube
Multi Agent	Cooperative Teams	▶️ Watch Travel Demo on YouTube
	Competitive Teams	▶️ Watch Debate Arena on YouTube
	Mixed of both Teams	Coming Soon 🚀

Contributing

We warmly welcome developers to join us in building and improving AWorld! Whether you're interested in enhancing the framework, fixing bugs, or adding new features, your contributions are valuable to us.

For academic citations or wish to contact us, please use the following BibTeX entry:

@software{aworld2025,
  author = {Agent Team at InclusionAI},
  title = {AWorld: A Framework for Agent Learning of Complex Tasks via Action-Observation-Reward Experience},
  year = {2025},
  url = {https://github.com/inclusionAI/AWorld},
  version = {0.1.0},
  publisher = {GitHub},
  email = {chenyi.zcy at antgroup.com}
}

License

This project is licensed under the MIT License - see the LICENSE file for details.