iod/AgentCoord
9
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Merge branch 'main' of https://github.com/AgentCoord/AgentCoord

Browse Source
This commit is contained in:
N-Believe 2024-04-08 14:05:41 +08:00
commit 631aa6dcdc
2 changed files with 12 additions and 29 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
README.md
View File

@ -56,7 +56,7 @@ You can set the configuration in ./backend/config/config.yaml. See [LLM configur
Currently, we support config agents by [role-prompting](https://arxiv.org/abs/2305.14688). You can customize your agents by changing the role prompts in AgentRepo\agentBoard_v1.json. We plan to support more methods to customize agents (e.g., supporting RAG, or providing a unified wrapper for customized agents) in the future.
## More Papers & Projects for LLM-based Multi-Agent Collaboration
If you’re interested in LLM-based multi-agent collaboration and want more papers & projects for reference, you may check out the [corpus](https://docs.google.com/spreadsheets/d/1HSl4AqIVXhUjZh0pRhz-brzfA7evh20Q/edit?usp=sharing&ouid=112400145401551512954&rtpof=true&sd=true) collected by us:
If you’re interested in LLM-based multi-agent collaboration and want more papers & projects for reference, you may check out the [corpus](https://docs.google.com/spreadsheets/d/1HSl4AqIVXhUjZh0pRhz-brzfA7evh20Q/edit?usp=sharing&ouid=112400145401551512954&rtpof=true&sd=true) collected by us. Any contribution to the corpus is also welcome.

39
backend/README.md
View File

@ -1,38 +1,21 @@
# AgentCoord: Visually Exploring Coordination Strategy for LLM-based Multi-Agent Collaboration
<p align="center">
<a ><img src="https://github.com/bopan3/AgentCoord_Backend/assets/21981916/bbad2f66-368f-488a-af36-72e79fdb6805" alt=" AgentCoord: Visually Exploring Coordination Strategy for LLM-based Multi-Agent Collaboration" width="200px"></a>
</p>
AgentCoord is an experimental open-source system to help general users design coordination strategies for multiple LLM-based agents (Research paper forthcoming).
# AgentCoord Backend
This is the backend for the AgentCoord project. Root project is located [here](https://github.com/AgentCoord/AgentCoord). See [LLM configuration](README.md#llm-configuration) for LLM configuration before you launch the backend server.
## System Usage
<a href="https://youtu.be/s56rHJx-eqY" target="_blank"><img src="https://github.com/bopan3/AgentCoord_Backend/assets/21981916/0d907e64-2a25-4bdf-977d-e90197ab1aab"
alt="System Usage Video" width="800" border="5" /></a>
## Launch
## Installation
### Installation
You can launch the backend by simply using `docker-compose` in the root directory of the project.
Or, you can launch the backend manually by following the steps below.
1. Install the dependencies by running `pip install -r requirements.txt`.
```bash
git clone https://github.com/AgentCoord/AgentCoord.git
cd AgentCoord
pip install -r requirements.txt
pip install -r ./requirements.txt
```
### Configuration
#### LLM configuration
You can set the configuration (i.e. API base, API key, Model name, Max tokens, Response per minute) for default LLM in config\config.yaml. Currently, we only support OpenAI’s LLMs as the default model. We recommend using gpt-4-0125-preview as the default model (WARNING: the execution process of multiple agents may consume a significant number of tokens).
You can switch to a fast mode that uses the Mistral 8×7B model with hardware acceleration by [Groq](https://groq.com/) for the first time in strategy generation to strike a balance of response quality and efficiency. To achieve this, you need to set the FAST_DESIGN_MODE field in the yaml file as True and fill the GROQ_API_KEY field with the api key of [Groq](https://wow.groq.com/).
#### Agent configuration
Currently, we support config agents by [role-prompting](https://arxiv.org/abs/2305.14688). You can customize your agents by changing the role prompts in AgentRepo\agentBoard_v1.json. We plan to support more methods to customize agents (e.g., supporting RAG, or providing a unified wrapper for customized agents).
### Launch
Execute the following command to launch the system:
2. launch the backend server (the default port is 8017)
```bash
python server.py
python ./server.py --port 8017
```
## More Papers & Projects for LLM-based Multi-Agent Collaboration
If you’re interested in LLM-based multi-agent collaboration and want more papers & projects for reference, you may check out the corpus collected by us: