Lightweight package for interacting with ChatGPT's API by OpenAI. Uses reverse engineered official API.

ChatGPT

Reverse Engineered ChatGPT by OpenAI. Extensible for chatbots etc.

Disclaimer

This is not an official OpenAI product. This is a personal project and is not affiliated with OpenAI in any way. Don't sue me

This is a library and not intended for direct CLI use

The CLI functionality is for demo and testing only.

@rawandahmad698 has a much better CLI tool at

PyChatGPT

Urgent help needed

• Writing tests
• More verbose error handling
• Decrecate bs4 in favor of pure requests

Wiki is now open to all contributors

Features

• No moderation
• Programmable

Setup

Install

pip3 install revChatGPT --upgrade

Email and password authentication

{
    "email": "<YOUR_EMAIL>",
    "password": "<YOUR_PASSWORD>"
}

Save this in config.json in current working directory

Running

 $ python3 -m revChatGPT            

    ChatGPT - A command-line interface to OpenAI's ChatGPT (https://chat.openai.com/chat)
    Repo: github.com/acheong08/ChatGPT
    Arguments: Use --stream to enable data streaming. (Doesn't work on MacOS)
    
Type '!help' to show commands
Press enter twice to submit your question.

You: !help


                !help - Show this message
                !reset - Forget the current conversation
                !refresh - Refresh the session authentication
                !exit - Exit the program

Refresh every so often in case the token expires.

Development:

pip3 install revChatGPT --upgrade

from revChatGPT.revChatGPT import Chatbot
import json

# Get your config in JSON
config = {
    "email": "<YOUR_EMAIL>",
    "password": "<YOUR_PASSWORD>"
}

chatbot = Chatbot(config, conversation_id=None)
chatbot.reset_chat() # Forgets conversation
chatbot.refresh_session() # Uses the session_token to get a new bearer token
resp = chatbot.get_chat_response(prompt, output="text") # Sends a request to the API and returns the response by OpenAI
resp['message'] # The message sent by the response
resp['conversation_id'] # The current conversation id
resp['parent_id'] # The ID of the response

 # This returns a stream of text (live update)
resp = chatbot.get_chat_response(prompt, output="stream") 
for message in resp: # You have to loop through the response stream
        print(line['message']) # Same format as text return type
        ...

This can be imported to projects for bots and much more. You can have multiple independent conversations by keeping track of the conversation_id.

Awesome ChatGPT

My list

If you have a cool project you want added to the list, open an issue.

Star History