OpenAI Chat Completion API Python-Flask Example

Streamy.CLI.Demo.Aug.2024.mp4

This repository demonstrates a local client-server implementation of the OpenAI Chat Completion API, utilizing the powerful GPT-4o language model to create conversational AI applications. The example features a Python Flask server that interfaces with the OpenAI API and a Python client that communicates with the server to facilitate the conversation.

About

The OpenAI Chat Completion API enables easy integration of AI-powered conversational capabilities into applications. The API uses the GPT-4o language model to understand and generate human-like text based on user input.

This project consists of two main components:

• Server (./server): A Flask application that processes incoming messages from the client, sends them to the OpenAI API, and returns the generated responses. It handles session state and conversation logging. You have two server options:

  • Default Chat: Standard chat mode where the entire response is sent at once.
  • Streaming Chat: Receives chunks of text to simulate dynamic typing, providing a more interactive experience.

• Client (./client): A command-line interface (CLI) that allows users to send messages to the Flask server and receive responses from the AI model. You can choose between:

  • Default Chat Client: For standard chat interactions.
  • Streaming Chat Client: For an experience where responses appear as if they are being typed out dynamically.

For more information about the OpenAI Chat Completion API, visit:

• Playground
• Chat Completions API Guide
• API Reference

Features

• Interactive CLI for sending and receiving messages.
• Flask server for API requests and response handling.
• Support for both default and stream chat modes.
• Conversation persistence on both client and server sides.
• Customizable user identification for conversation logging.

Authentication

The script requires an API key from OpenAI for authentication. Before running the script, set up an environment variable OPENAI_API_KEY with your OpenAI API key. Retrieve your API key from your API Keys page on OpenAI.

For assistance on setting up environment variables:

• On Windows, follow Microsoft's guide on Environment Variables.
• On Linux/Unix, refer to the guide on setting environment variables in Linux.

Setup

To get started with this example:

1. Clone the repository to your local machine.

2. Install the required dependencies by running pip install -r requirements.txt in both the ./client and ./server directories.

3. Set your OpenAI API key as an environment variable on your PC:

   • On Windows Command Prompt:

     setx OPENAI_API_KEY "your_api_key_here"

   • On Windows PowerShell:

     [System.Environment]::SetEnvironmentVariable('OPENAI_API_KEY', 'your_api_key_here', 'User')

   • On macOS/Linux:

     export OPENAI_API_KEY='your_api_key_here'

4. Choose your server and client scripts:

   • Default Chat:
     • Run the server with python flask_default_chat.py within the /server directory.
     • Run the client with python default_chat.py within the /client directory.
   • Stream Chat:
     • Run the server with python flask_stream_chat.py within the /server directory.
     • Run the client with python stream_chat.py within the /client directory.

5. Follow the prompts in the client to begin a conversation with the AI.

6. Chat history files will be saved to the respective ./client/chat_history and ./server/chat_history directories.

Contributions

Contributions to this example are welcome! If you have an improvement or encounter an issue, please feel free to open an issue or submit a pull request.

License

This project is open-sourced under the MIT License. See the LICENSE file for more details.