ChatGPT

ChatGPT Project Overview

The ChatGPT project provides a free reverse proxy API for accessing OpenAI's gpt-3.5-turbo model. This setup allows users to integrate AI functionalities without modifying their existing code. Although the project is currently outdated and needs updating, users can still benefit from it using hosted models.

Key Features

• Streaming Responses: The API delivers responses as soon as they're ready, ensuring real-time interaction.
• OpenAI Compatibility: It aligns with OpenAI's official API, allowing seamless integration.
• Free Access: No API key is required, making advanced AI accessible to everyone at no cost.

Installation and Self-Hosting

There are multiple methods to set up the API:

Using Docker

1. Install Docker by following the Docker Installation Docs.
2. Execute the command:

   docker run -dp 3040:3040 pawanosman/chatgpt:latest
   

3. Access the local server's API at http://localhost:3040/v1/chat/completions.

On Your PC/Server

This method may need a VPN for unsupported countries:

1. Install NodeJs (v19+): Download NodeJs.
2. Clone the repository:

   git clone https://github.com/PawanOsman/ChatGPT.git
   

3. Run start.bat (Windows) or bash start.sh (Linux) to start the server.
4. Access the API at http://localhost:3040/v1/chat/completions.

On Android via Termux

1. Install Termux.
2. Update and upgrade packages:

   apt update && apt upgrade
   

3. Install the necessary tools:

   apt install -y git nodejs
   

4. Clone and run the project:

   git clone https://github.com/PawanOsman/ChatGPT.git
   cd ChatGPT
   bash start.sh
   

5. Access the API at http://localhost:3040/v1/chat/completions.

Using Hosted API

Join their Discord server to access the hosted API:

1. Obtain an API key from the #Bot channel.
2. Use the key to make requests to https://api.pawan.krd/v1/chat/completions.

Usage Examples

Python

import openai

openai.api_key = 'anything'
openai.base_url = "http://localhost:3040/v1/"

completion = openai.chat.completions.create(
    model="gpt-3.5-turbo",
    messages=[{"role": "user", "content": "How do I list all files in a directory using Python?"}],
)

print(completion.choices[0].message.content)

Node.js

import OpenAI from 'openai';

const openai = new OpenAI({
	apiKey: "anything",
	baseURL: "http://localhost:3040/v1",
});

const chatCompletion = await openai.chat.completions.create({
  messages: [{ role: 'user', content: 'Say this is a test' }],
  model: 'gpt-3.5-turbo',
});

console.log(chatCompletion.choices[0].message.content);

License

The project is available under the AGPL-3.0 License, providing open access to the underlying code.