PantheonAI is a multi-user long-term history language learning model (LLM) chat application. It uses OpenAI's GPT-3.5-turbo model to generate responses to user prompts and maintains a history of the conversation.
- Generate responses to user prompts using GPT-3.5-turbo.
- Maintain a history of the conversation.
- Support for multiple prompts.
- A client interface for interacting with the chat application.
This project uses Poetry for dependency management. To install the project, follow these steps:
- Clone the repository:
git clone https://github.com/mehraneftekhari99/PantheonAI.git- Navigate to the project directory:
cd PantheonAI- Install the dependencies:
poetry installTo run the project, use the following command:
python run.pyThe application will start on localhost port 5000.
To interact with the chat application, run the client script:
python client.pyThe client will connect to the server running on localhost port 5000 by default. You can change the server IP and port by setting the PANTHEON_SERVER_IP and PANTHEON_SERVER_PORT environment variables, respectively.
This project depends on the following Python packages:
- zep-python
- flask
- openai
- colorama
- gnureadline
- fastapi
- uvicorn
- gunicorn
- pydantic
Note: Zep should be running on localhost. You can use the official docker compose.
Contributions are welcome. Please open an issue to discuss your ideas before making changes.
This project is licensed under the GPL-3.0 License. City names list is taken from simplemaps with some minor modifications.