A web application that allows users to interact with Large Language Model and recognize the importance of privacy protection in large language model. You can experience the web frontend here: https://artessay.github.io/GAN-LLM/ .
- User-friendly interface for making requests to the Large Language Model
- Responses are displayed in a chat-like format
- Select Models (GPT, GAN-LLM) based on your needs
- Highlight code syntax
- For client, I used React.
- For server, I used Flask.
This guide will help you set up the repository on your local machine. Please follow these steps carefully to ensure a smooth setup process.
Use the following command to clone the repository:
git clone https://github.com/Artessay/GAN-LLM.git
- Navigate to server directory
cd server # Navigate to the server directory:
- Install dependencies
pip install -r requirements.txt #install the backend dependencies
- Start the backend server by running the following command:
python app.py
- Navigate to the client directory:
cd client
- Run the following command to install the frontend dependencies:
npm install
- Set the
REACT_APP_BACKEND_URL
in the.env
file to the URL of your backend server. For local development, use the following URL. Note: if you want to deploy it on the website, you should use your public IP address.
REACT_APP_BACKEND_URL=http://localhost:3080/
- Start the frontend app by running the following command:
npm start
If you wish to host both the backend and frontend on the same port/URL, follow these steps:
- Build the frontend by running the following command in the
client
directory:
npm run build
-
Copy the
build
directory to theserver
directory and rename it tofrontend
. -
Start the backend server using the instructions in the "Backend Setup" section.
-
Once the setup process is complete, the frontend will be accessible at the URL of your backend server.
- Type in the input field and press enter or click on the send button to make a request to the Large Language Model
- Use control+enter to add line breaks in the input field
- Responses are displayed in the chat-like format on top of the page
- Generate code, including translating natural language to code
This project welcomes contributions and suggestions for improvements. If you have any ideas, please feel free to open an issue or create a pull request.
Thank you for your consideration.