https://github.com/agiresearch/OpenAGI
OpenAGI: When LLM Meets Domain Experts
https://github.com/agiresearch/OpenAGI
agi llm opensource
Last synced: 18 days ago
JSON representation
OpenAGI: When LLM Meets Domain Experts
- Host: GitHub
- URL: https://github.com/agiresearch/OpenAGI
- Owner: agiresearch
- License: mit
- Created: 2023-04-08T02:58:55.000Z (about 2 years ago)
- Default Branch: main
- Last Pushed: 2024-09-02T19:40:36.000Z (7 months ago)
- Last Synced: 2024-10-29T15:05:25.821Z (6 months ago)
- Topics: agi, llm, opensource
- Language: Python
- Homepage: http://aios.foundation
- Size: 330 MB
- Stars: 1,946
- Watchers: 28
- Forks: 162
- Open Issues: 5
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
- Awesome_Multimodel_LLM - OpenAGI - When LLM Meets Domain Experts. (Other useful resources)
- awesome-llm - OpenAGI - 当LLM遇到领域专家时。 (其他资源 / LLM 评估工具)
- awesome-llm - OpenAGI - 当LLM遇到领域专家时。 (其他资源 / LLM 评估工具)
- Awesome-AGI - OpenAGI - | (Frameworks and Platforms)
- Awesome-LLM - OpenAGI - When LLM Meets Domain Experts. (Miscellaneous)
- Awesome-AI-Agents - OpenAGI - "May the Force be with LLM and Domain Experts."  (Platforms/API / Advanced Components)
- Awesome-AI-Agents - OpenAGI - "May the Force be with LLM and Domain Experts."  (Platforms/API / Advanced Components)
- awesome-production-machine-learning - OpenAGI - OpenAGI is used as the agent creation package to build agents for AIOS. (Agentic Framework)
- awesome_ai_agents - OpenAGI - OpenAGI by AGI Research is an open-source platform integrating Large Language Models (LLMs) with domain-specific expert models for complex task-solving, fostering a paradigm where LLMs operate various external models, accompanied by a Reinforcement Learning from Task Feedback (RLTF) mechanism for self-improvement [github](https://github.com/agiresearch/OpenAGI) | [github profile](https://github.com/agiresearch) (Learning / Repositories)
- awesome_ai_agents - OpenAGI - OpenAGI by AGI Research is an open-source platform integrating Large Language Models (LLMs) with domain-specific expert models for complex task-solving, fostering a paradigm where LLMs operate various external models, accompanied by a Reinforcement Learning from Task Feedback (RLTF) mechanism for self-improvement [github](https://github.com/agiresearch/OpenAGI) | [github profile](https://github.com/agiresearch) (Learning / Repositories)
README
# OpenAGI: Package for AI Agent Creation
[](https://github.com/agiresearch/OpenAGI/blob/main/LICENSE)
## ✈️ Getting Started
OpenAGI is used as the agent creation package to build agents for [AIOS](https://github.com/agiresearch/AIOS).
**Notice:** For building up agents in the AIOS, please migrate to the [Cerebrum](https://github.com/agiresearch/Cerebrum), which is our latest SDK to connect with AIOS kernel.
### Installation
From PyPI
```
pip install pyopenagi
```
Locally
```
git clone https://agiresearch/OpenAGI
cd OpenAGI
pip install -e .
```
### Usage
#### Add a new agent
To add a new agent, first you need to create a folder under the pyopenagi/agents folder.
The folder needs to be the following structure:
```
- pyopenagi/agents
- author
- agent_name
- agent.py # main code for the agent execution logic
- config.json # set up configurations for agent
- meta_requirements.txt # dependencies that the agent needs
```
If you want to use external tools provided by openagi in your agents, you can follow instructions of setting up tools in [How to setup external tools](./tools.md).
If you want to add new tools for your developing agent,
you need to add a new tool file in the [folder](./pyopenagi/tools/).
#### Upload agent
If you have developed and tested your agent, and you would like to share your agents, you can use the following to upload your agents
```
python pyopenagi/agents/interact.py --mode upload --agent
```
💡Note that the `agent` param must exactly match the folder you put your agent locally.
#### Download agent
If you want to look at implementations of other agents that others have developed, you can use the following command:
```
python pyopenagi/agents/interact.py --mode download --agent
```
## 🚀 Contributions
For detailed information on how to contribute, see [CONTRIBUTE](./CONTRIBUTE.md). If you would like to contribute to the codebase, [issues](https://github.com/agiresearch/OpenAGI/issues) or [pull requests](https://github.com/agiresearch/OpenAGI/pulls) are always welcome!
## 🖋️ Research
Please check out our [implementation](https://github.com/agiresearch/OpenAGI/tree/research) for our research paper [OpenAGI: When LLM Meets Domain Experts](https://arxiv.org/abs/2304.04370).
```
@article{openagi,
title={OpenAGI: When LLM Meets Domain Experts},
author={Ge, Yingqiang and Hua, Wenyue and Mei, Kai and Ji, Jianchao and Tan, Juntao and Xu, Shuyuan and Li, Zelong and Zhang, Yongfeng},
journal={In Advances in Neural Information Processing Systems (NeurIPS)},
year={2023}
}
```
## 🌍 OpenAGI Contributors
[](https://github.com/agiresearch/OpenAGI/graphs/contributors)
## 🌟 Star History
[](https://star-history.com/#agiresearch/OpenAGI&Date)