Assign different roles to GPTs to form a collaborative entity for complex tasks.
🚀 Feb. 19, 2025: Today we are officially launching our natural language programming product: MGX (MetaGPT X) - the world's first AI agent development team. Offical website Twitter
🚀 Feb. 17, 2025: We introduced two papers: SPO and AOT, check the code!
🚀 Jan. 22, 2025: Our paper AFlow: Automating Agentic Workflow Generation accepted for oral presentation (top 1.8%) at ICLR 2025, ranking #2 in the LLM-based Agent category.
🚀 Oct. 29, 2024: We introduced three papers: AFLOW, FACT, and SELA, check the code!
🚀 Mar. 29, 2024: v0.8.0 released. Now you can use Data Interpreter (arxiv, example, code) via pypi package import. Meanwhile, we integrated the RAG module and supported multiple new LLMs.
🚀 Feb. 08, 2024: v0.7.0 released, supporting assigning different LLMs to different Roles. We also introduced Data Interpreter, a powerful agent capable of solving a wide range of real-world problems.
🚀 Jan. 16, 2024: Our paper MetaGPT: Meta Programming for A Multi-Agent Collaborative Framework accepted for oral presentation (top 1.2%) at ICLR 2024, ranking #1 in the LLM-based Agent category.
🚀 Jan. 03, 2024: v0.6.0 released, new features include serialization, upgraded OpenAI package and supported multiple LLM, provided minimal example for debate etc.
🚀 Dec. 15, 2023: v0.5.0 released, introducing some experimental features such as incremental development, multilingual, multiple programming languages, etc.
🔥 Nov. 08, 2023: MetaGPT is selected into Open100: Top 100 Open Source achievements.
🔥 Sep. 01, 2023: MetaGPT tops GitHub Trending Monthly for the 17th time in August 2023.
🌟 Jun. 30, 2023: MetaGPT is now open source.
🌟 Apr. 24, 2023: First line of MetaGPT code committed.
Code = SOP(Team) is the core philosophy. We materialize SOP and apply it to teams composed of LLMs.
Software Company Multi-Agent Schematic (Gradually Implementing)
Ensure that Python 3.9 or later, but less than 3.12, is installed on your system. You can check this by using:
python --version.
You can use conda like this:conda create -n metagpt python=3.9 && conda activate metagpt
pip install --upgrade metagpt
# or `pip install --upgrade git+https://github.com/geekan/MetaGPT.git`
# or `git clone https://github.com/geekan/MetaGPT && cd MetaGPT && pip install --upgrade -e .`
For detailed installation guidance, please refer to cli_install or docker_install
You can init the config of MetaGPT by running the following command, or manually create ~/.metagpt/config2.yaml file:
# Check https://docs.deepwisdom.ai/main/en/guide/get_started/configuration.html for more details
metagpt --init-config # it will create ~/.metagpt/config2.yaml, just modify it to your needs
You can configure ~/.metagpt/config2.yaml according to the example and doc:
llm:
api_type: "openai" # or azure / ollama / groq etc. Check LLMType for more options
model: "gpt-4-turbo" # or gpt-3.5-turbo
base_url: "https://api.openai.com/v1" # or forward url / other llm url
api_key: "YOUR_API_KEY"
After installation, you can use MetaGPT at CLI
metagpt "Create a 2048 game" # this will create a repo in ./workspace
or use it as library
from metagpt.software_company import generate_repo, ProjectRepo
repo: ProjectRepo = generate_repo("Create a 2048 game") # or ProjectRepo("<path>")
print(repo) # it will print the repo structure with files
You can also use Data Interpreter to write code:
import asyncio
from metagpt.roles.di.data_interpreter import DataInterpreter
async def main():
di = DataInterpreter()
await di.run("Run data analysis on sklearn Iris dataset, include a plot")
asyncio.run(main()) # or await main() in a jupyter notebook setting
https://github.com/geekan/MetaGPT/assets/34952977/34345016-5d13-489d-b9f9-b82ace413419
📢 Join Our Discord Channel! Looking forward to seeing you there! 🎉
📝 Fill out the form to become a contributor. We are looking forward to your participation!
If you have any questions or feedback about this project, please feel free to contact us. We highly appreciate your suggestions!
We will respond to all questions within 2-3 business days.
To stay updated with the latest research and development, follow @MetaGPT_ on Twitter.
To cite MetaGPT or Data Interpreter in publications, please use the following BibTeX entries.
@inproceedings{hong2024metagpt,
title={Meta{GPT}: Meta Programming for A Multi-Agent Collaborative Framework},
author={Sirui Hong and Mingchen Zhuge and Jonathan Chen and Xiawu Zheng and Yuheng Cheng and Jinlin Wang and Ceyao Zhang and Zili Wang and Steven Ka Shing Yau and Zijuan Lin and Liyang Zhou and Chenyu Ran and Lingfeng Xiao and Chenglin Wu and J{\"u}rgen Schmidhuber},
booktitle={The Twelfth International Conference on Learning Representations},
year={2024},
url={https://openreview.net/forum?id=VtmBAGCN7o}
}
@misc{teng2025atom,
title={Atom of Thoughts for Markov LLM Test-Time Scaling},
author={Fengwei Teng and Zhaoyang Yu and Quan Shi and Jiayi Zhang and Chenglin Wu and Yuyu Luo},
year={2025},
eprint={2502.12018},
archivePrefix={arXiv},
primaryClass={cs.CL},
url={https://arxiv.org/abs/2502.12018},
}
@misc{xiang2025self,
title={Self-Supervised Prompt Optimization},
author={Jinyu Xiang and Jiayi Zhang and Zhaoyang Yu and Fengwei Teng and Jinhao Tu and Xinbing Liang and Sirui Hong and Chenglin Wu and Yuyu Luo},
year={2025},
eprint={2502.06855},
archivePrefix={arXiv},
primaryClass={cs.CL},
url={https://arxiv.org/abs/2502.06855},
}
@inproceedings{wang2025fact,
title={FACT: Examining the Effectiveness of Iterative Context Rewriting for Multi-fact Retrieval},
author={Jinlin Wang and Suyuchen Wang and Ziwen Xia and Sirui Hong and Yun Zhu and Bang Liu and Chenglin Wu},
booktitle={The 2025 Annual Conference of the Nations of the Americas Chapter of the ACL},
year={2025},
url={https://openreview.net/forum?id=VXOircx5h3}
}
@misc{chi2024sela,
title={SELA: Tree-Search Enhanced LLM Agents for Automated Machine Learning},
author={Yizhou Chi and Yizhang Lin and Sirui Hong and Duyi Pan and Yaying Fei and Guanghao Mei and Bangbang Liu and Tianqi Pang and Jacky Kwok and Ceyao Zhang and Bang Liu and Chenglin Wu},
year={2024},
eprint={2410.17238},
archivePrefix={arXiv},
primaryClass={cs.AI},
url={https://arxiv.org/abs/2410.17238},
}
@inproceedings{zhang2025aflow,
title={{AF}low: Automating Agentic Workflow Generation},
author={Jiayi Zhang and Jinyu Xiang and Zhaoyang Yu and Fengwei Teng and Xiong-Hui Chen and Jiaqi Chen and Mingchen Zhuge and Xin Cheng and Sirui Hong and Jinlin Wang and Bingnan Zheng and Bang Liu and Yuyu Luo and Chenglin Wu},
booktitle={The Thirteenth International Conference on Learning Representations},
year={2025},
url={https://openreview.net/forum?id=z5uVAKwmjf}
}
@misc{hong2024data,
title={Data Interpreter: An LLM Agent For Data Science},
author={Sirui Hong and Yizhang Lin and Bang Liu and Bangbang Liu and Binhao Wu and Danyang Li and Jiaqi Chen and Jiayi Zhang and Jinlin Wang and Li Zhang and Lingyao Zhang and Min Yang and Mingchen Zhuge and Taicheng Guo and Tuo Zhou and Wei Tao and Wenyi Wang and Xiangru Tang and Xiangtao Lu and Xiawu Zheng and Xinbing Liang and Yaying Fei and Yuheng Cheng and Zongze Xu and Chenglin Wu},
year={2024},
eprint={2402.18679},
archivePrefix={arXiv},
primaryClass={cs.AI},
url={https://arxiv.org/abs/2402.18679},
}
FAQs
The Multi-Agent Framework
We found that metagpt demonstrated a healthy version release cadence and project activity because the last version was released less than a year ago. It has 1 open source maintainer collaborating on the project.
Did you know?
Socket for GitHub automatically highlights issues in each pull request and monitors the health of all your open source dependencies. Discover the contents of your packages and block harmful activity before you install or update your dependencies.
Research
/Security News
A flawed sandbox in @nestjs/devtools-integration lets attackers run code on your machine via CSRF, leading to full Remote Code Execution (RCE).
Product
Customize license detection with Socket’s new license overlays: gain control, reduce noise, and handle edge cases with precision.
Product
Socket now supports Rust and Cargo, offering package search for all users and experimental SBOM generation for enterprise projects.