This is a simple Flask application that allows users to chat with GPT-3 using the text-davinci-003 model. The application has a chat page that takes user input and generates a response from GPT-3, and it also stores the conversation data in a file for later analysis.
The code uses the OpenAI Python library to interact with the GPT-3 model and generate responses.
To use the application, you will need to set up your OpenAI API key and install the required dependencies. See the instructions below for details.
-
Clone the repository and navigate to the project directory:
git clone https://github.com/amcmurchie/gpt3-chat.git cd gpt3-chat -
Create a file called
.KEY.txtin the project directory and add your OpenAI API key to the file. -
Install the required dependencies using
pip:pip install -r requirements.txt -
Run the Flask app using the following command:
python app.py -
Open
http://127.0.0.1:5000/chatin your web browser to access the chat page.
In addition to the Flask app, this code also includes a terminal-only version that can be run by calling the conversation() function directly, without using the Flask app or the /conversation route. This allows you to test the conversation functionality without running the web application.
To use the terminal-only version, you can run the following command in the project directory:
python -c "from app import conversation; conversation()"This will start a conversation with GPT-3 and print the responses to the terminal. The conversation data will also be written to the conversation.txt file in the project directory.