GPTeam: Collaborative AI Agents

Aug 25, 2021 | Educational




Web App
Blog Post

GPTeam

About GPTeam

GPTeam uses GPT-4 to create multiple agents who collaborate to achieve predefined goals. The main objective of this project is to explore the potential of GPT models in enhancing multi-agent productivity and effective communication.

See a video demo here.

Read more about the architecture here.

Getting Started

To begin exploring GPTeam, follow these steps:

  1. Clone the project repository to your local machine.
  2. Move to the repository: cd gpteam.
  3. Run python setup.py to check your environment setup and configure it as needed.
  4. Update the environment variables in .env with your API Keys.
  5. Launch the world by running poetry run world.

To run the world cheaply, you can use poetry run world --turbo. This will use gpt3.5-turbo for all LLM calls which is a lot cheaper, but expect worse results!

Now you can observe the world in action and watch as the agents interact with each other, working together to accomplish their assigned directives.

How It Works

GPTeam employs separate agents, each equipped with memory, that interact with one another using communication as a tool. Imagine a group of chefs in a large kitchen, each with their own ingredients and tools (the memory). They might discuss which dish to prepare (communication), then go about their tasks in different sections of the kitchen (locations) while occasionally checking in on one another’s progress. This collaborative atmosphere allows them to prepare a multi-course meal more efficiently than if they were each working in isolation!

The implementation of agent memory and reflection is inspired by this research paper. Agents can perform tasks in parallel towards common goals, effectively creating a bustling and collaborative environment.

Viewing Agents

The world is a busy place! To get a view of what different agents are doing while the world is running, you can visit the agents folder where there is a text file for each agent containing a summary of their current state.

Changing the World

To change the world, all you need to do is:

  1. Make changes to the config.json by updating the available agents or locations.
  2. Reset your database: poetry run db-reset.
  3. Run the world again: poetry run world.

Setting Up the Discord Integration

Read through the dedicated Discord setup docs.

Using with Anthropic Claude

Make sure you have an ANTHROPIC_API_KEY in your environment variables. Then you can use poetry run world --claude which will run the world using claude-v1 for some calls and claude-v1-instant for others.

Using with Window

Make sure you have the Window extension installed, then you can use poetry run world --window. Some models may be slow to respond, since the prompts are very long.

Contributing

We enthusiastically welcome contributions to GPTeam! To contribute, please follow these steps:

  1. Fork the project repository to your own account.
  2. Create a new branch for your changes.
  3. Implement your changes to the project code.
  4. Submit a pull request to the main project repository.

We will review your pull request and provide feedback as necessary.

Troubleshooting

If you encounter any issues during setup or while running GPTeam, here are some troubleshooting tips:

  • Ensure your API keys are correct and have the necessary permissions.
  • Double-check if you have all required dependencies installed.
  • If the world doesn’t launch, try resetting your database as described above.
  • Make sure you are in the correct directory when running commands.

For more insights, updates, or to collaborate on AI development projects, stay connected with fxis.ai.

At fxis.ai, we believe that such advancements are crucial for the future of AI, as they enable more comprehensive and effective solutions. Our team is continually exploring new methodologies to push the envelope in artificial intelligence, ensuring that our clients benefit from the latest technological innovations.

Stay Informed with the Newest F(x) Insights and Blogs

Tech News and Blog Highlights, Straight to Your Inbox