Formatting and consistency is extremely important in large projects. However, as it is currently exam season for me, I don't have enough time to dedicate to this.

If anyone is willing, please format the code so that it passes PyLint. I'm not sure if this can be done automatically.

Sincere thanks

Hi Great project and awesome work, by any chance you can add proxy for request,just like we can pass in

config = {
    "email": "xxx",
    "password": "",
    "https": ""
}

for request.

Thx again

here is my code, The first two get_chat_response is no problem, but I reset the conversation_id then call get_chat_response it will abort that Too many requests, please slow down

chatbot = Chatbot(config, conversation_id=None)
chatbot.refresh_session() 
# 1
resp = chatbot.get_chat_response('what is python')
print(resp)
# 2
resp = chatbot.get_chat_response('continue') 
print(resp)

# 3
id = resp['conversation_id']
chatbot = Chatbot(config, conversation_id=id)
resp = chatbot.get_chat_response('continue') 
print(resp)

Hi,

I am writing a Telegram bot via python-telegram-bot to interact with ChatGPT. I have defined some env vars i.e. CHATGPT_AUTH_KEY (this I got from /beta) and CHATGPT_SESSION_TOKEN as the tutorial points out. This is the code:

# ChatGPT
chatgpt = Chatbot({"Authorization":ENVIRONMENT.CHATGPT_AUTH_KEY, "session_token": ENVIRONMENT.CHATGPT_SESSION_TOKEN}, conversation_id=None)
chatgpt.reset_chat()

However when I refresh the session chatgpt.refresh_session()I get:

Error refreshing session
{}

So I inspected the chatgpt.headers and that gives:

{'Accept': 'application/json',
 'Authorization': 'Bearer sk-WpVb8p90gjkYwwglGlYyT3BlbkFJ6eRims************',
 'Content-Type': 'application/json'}

There's no 'session_token' in there. It looks Chatbot is not able to grab the session_token header and related value. I tried to supply also empty Authorization header to see if it was something related to passing configuration but in that case I got:

{'Accept': 'application/json',
 'Authorization': 'Bearer ',
 'Content-Type': 'application/json'}

I triend also to chatgpt.refresh_headers but that still does not fix the problem.

Do you have any suggestions?

The ChatGPT is blocked. I wish that would be open source in the near future.

Plus I'd like to assure you, they can't, Code from my repo has been updated with the new headers.
https://github.com/rawandahmad698/PyChatGPT

title lol

Attempting to install latest version gives an error when installing the lxml dependency on Windows 11

Attempted to install lxml NuGet package in Visual Studio but still can't build this package. There is no wheel package for Windows Python 3.11

PS Z:\Programming\chatGPT\Playground> pip3 install revChatGPT --upgrade
Requirement already satisfied: revChatGPT in c:\python311\lib\site-packages (0.0.14)
Collecting revChatGPT
  Downloading revChatGPT-0.0.23.5-py3-none-any.whl (14 kB)
Collecting bs4
  Downloading bs4-0.0.1.tar.gz (1.1 kB)
  Preparing metadata (setup.py) ... done
Collecting lxml
  Downloading lxml-4.9.1.tar.gz (3.4 MB)
     ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 3.4/3.4 MB 54.3 MB/s eta 0:00:00
  Preparing metadata (setup.py) ... done
. . .
Installing collected packages: tls-client, soupsieve, lxml, beautifulsoup4, bs4, revChatGPT
  DEPRECATION: lxml is being installed using the legacy 'setup.py install' method, because it does not have a 'pyproject.toml' and the 'wheel' package is not installed. pip 23.1 will enforce this behaviour change. A possible replacement is to enable the '--use-pep517' option. Discussion can be found at https://github.com/pypa/pip/issues/8559
  Running setup.py install for lxml ... error
  error: subprocess-exited-with-error

  × Running setup.py install for lxml did not run successfully.
  │ exit code: 1
  ╰─> [96 lines of output]
      Building lxml version 4.9.1.
      Building without Cython.
      Building against pre-built libxml2 andl libxslt libraries
      running install
      C:\Python311\Lib\site-packages\setuptools\command\install.py:34: SetuptoolsDeprecationWarning: setup.py install is deprecated. Use build and pip and other standards-based tools.
. . .
      "C:\Program Files\Microsoft Visual Studio\2022\Enterprise\VC\Tools\MSVC\14.33.31629\bin\HostX86\x64\cl.exe" /c /nologo /O2 /W3 /GL /DNDEBUG /MD -DCYTHON_CLINE_IN_TRACEBACK=0 -Isrc -Isrc\lxml\includes -IC:\Python311\include -IC:\Python311\Include "-IC:\Program Files\Microsoft Visual Studio\2022\Enterprise\VC\Tools\MSVC\14.33.31629\include" "-IC:\Program Files\Microsoft Visual Studio\2022\Enterprise\VC\Tools\MSVC\14.33.31629\ATLMFC\include" "-IC:\Program Files\Microsoft Visual Studio\2022\Enterprise\VC\Auxiliary\VS\include" "-IC:\Program Files (x86)\Windows Kits\10\include\10.0.19041.0\ucrt" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.19041.0\\um" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.19041.0\\shared" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.19041.0\\winrt" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.19041.0\\cppwinrt" "-IC:\Program Files (x86)\Windows Kits\NETFXSDK\4.8\include\um" /Tcsrc\lxml\etree.c /Fobuild\temp.win-amd64-cpython-311\Release\src\lxml\etree.obj -w
      cl : Command line warning D9025 : overriding '/W3' with '/w'
      etree.c
      C:\Users\user\AppData\Local\Temp\pip-install-rh5v03nk\lxml_23167b2f100d4470a19a380bfa8a641e\src\lxml\includes/etree_defs.h(14): fatal error C1083: Cannot open include file: 'libxml/xmlversion.h': No such file or directory
      Compile failed: command 'C:\\Program Files\\Microsoft Visual Studio\\2022\\Enterprise\\VC\\Tools\\MSVC\\14.33.31629\\bin\\HostX86\\x64\\cl.exe' failed with exit code 2
ocal\Temp\xmlXPathInitjtxtwcqk.c(1): fatal error C1083: Cannot open include file: 'libxml/xpath.h': No such file or directory
      error: command 'C:\\Program Files\\Microsoft Visual Studio\\2022\\Enterprise\\VC\\Tools\\MSVC\\14.33.31629\\bin\\HostX86\\x64\\cl.exe' failed with exit code 2
      *********************************************************************************
      Could not find function xmlCheckVersion in library libxml2. Is libxml2 installed?
      *********************************************************************************
      [end of output]

  note: This error originates from a subprocess, and is likely not a problem with pip.
error: legacy-install-failure

× Encountered error while trying to install package.
╰─> lxml

note: This is an issue with the package mentioned above, not pip.
hint: See above for output from the failure.

Describe the bug
Tried to refresh the chatbot

To Reproduce
Steps to reproduce the behavior:

1. Use this code:

from revChatGPT.revChatGPT import Chatbot
chatbot = Chatbot(config, conversation_id=None)
chatbot.refresh_session()
chatbot.reset_chat() # Forgets conversation

3. See error

Environment (please complete the following information):

• MacOS
• Python version: 3.8.10
• Version: 0.0.23.5

Additional context
I am using python 3.8 to be compatible with other libraries in this project

Logging in...
Captcha detected
Error refreshing session
can only concatenate str (not "NoneType") to str
...

Expected behavior for the difficulty in resolving captcha issue?

In your "Features" you mention

Uncensored
No moderation

can you please explain how it's more uncensored than the regular site?
Is there censoring on the frontend? If so, how can i trigger the censoring on the frontend?

By using the __Secure-next-auth.session-token cookie, you are able to generate access tokens just like a browser would. This would allow users to set a config value once and forget (until the session expires, which the API says ~1 month).

Since the __Secure-next-auth.session-token value changes after every session call, you would need to store it in a requests.Session, as well as dump the new cookie value into the config for preservation.

Approximate Example:

import json
from pathlib import Path
from requests import Session

config_path = Path('config.json')
config = json.loads(config_path.read_text())

session = Session()
session.headers |= {'Accept': 'application/json'}
session.cookies.set('__Secure-next-auth.session-token', config['Authorization'], domain='chat.openai.com')

def refresh_auth():
    response = session.get('https://chat.openai.com/api/auth/session')
    session.headers |= {'Authorization': f'Bearer {response.json()['accessToken']}'}
    config |= {'Authorization': session.cookies.get('__Secure-next-auth.session-token', domain='chat.openai.com')}
    config_path.write_text(json.dumps(config))

I have a working solution in my own project, though its not very clean and uses PySide6.QNetwork as opposed to requests.

When using the api, I encounter this error:

Error refreshing session

Exception in Tkinter callback
Traceback (most recent call last):
File "/Users/bytedance/PycharmProjects/ChineseAiDungeonChatGPT/venv/lib/python3.6/site-packages/revChatGPT/revChatGPT.py", line 61, in get_chat_text
response = response.text.splitlines()[-4]
IndexError: list index out of range

Now it seems that the use of api is detected and baned by openai. Can someone please confirm this? Or is there any way around it?

Hi!

I created a serverless Telegram bot for ChatGPT using AWS Lambda. The implementation is based on your reverse-engineering API. Credits to you in my README.md

https://github.com/franalgaba/chatgpt-telegram-bot-serverless

Thank you :)
Fran.

I can't conduct further testing until it is reinstated. If anyone can help, please do.

Describe the bug
python -m revChatGPT
report an error on my computer

Traceback (most recent call last):
  File "E:\python\lib\runpy.py", line 196, in _run_module_as_main
    return _run_code(code, main_globals, None,
  File "E:\python\lib\runpy.py", line 86, in _run_code
    exec(code, run_globals)
  File "E:\python\lib\site-packages\revChatGPT\__main__.py", line 1, in <module>
    from revChatGPT.revChatGPT import Chatbot
  File "E:\python\lib\site-packages\revChatGPT\revChatGPT.py", line 4, in <module>
    import tls_client
  File "E:\python\lib\site-packages\tls_client\__init__.py", line 15, in <module>
    from .sessions import Session
  File "E:\python\lib\site-packages\tls_client\sessions.py", line 1, in <module>
    from .cffi import request
  File "E:\python\lib\site-packages\tls_client\cffi.py", line 15, in <module>
    library = ctypes.cdll.LoadLibrary(f'{root_dir}/dependencies/tls-client{file_ext}')
  File "E:\python\lib\ctypes\__init__.py", line 452, in LoadLibrary
    return self._dlltype(name)
  File "E:\python\lib\ctypes\__init__.py", line 374, in __init__
    self._handle = _dlopen(self._name, mode)
OSError: [WinError 193] %1 Is not a valid Win32 application.

I'm not familiar with python, so I can't find out the specific problems
I hope to get help

Environmen:

• OS: Windows 10
• Python version: Python 3.10.4
• Version:

Name: revChatGPT
Version: 0.0.26.2
Summary: ChatGPT is a reverse engineering of OpenAI's ChatGPT API
Home-page: https://github.com/acheong08/ChatGPT
Author: Antonio Cheong
Author-email: [email protected]
License: GNU General Public License v2.0
Location: e:\python\lib\site-packages
Requires: bs4, lxml, requests, tls-client
Required-by:

Hello does not send the message completely after certain character length.

For example :

hello

hel

I run:

chatbot = Chatbot(config, conversation_id=None)
chatbot.reset_chat()
chatbot.refresh_session()
resp = chatbot.get_chat_response("Complete the following blog post:\n" + dataobj.content, output="text")
print(resp["message"])

and get:

TypeError: 'generator' object is not subscriptable

This is referring to the resp object.

Thanks for your help, and for making the project in general!

{"detail":{"message":"Your authentication token has expired. Please try signing in again.","type":"invalid_request_error","param":null,"code":"token_expired"}}
Response is not in the correct format

Please refresh your tokens

i think chatgpt answer should be split in multiple discord message there

I came across some limitations like the possibility to scale as big as needed etc. I ended up using the Menubar X app on Mac that is (exclusive to that platform). It allows for arbitrary apps to be placed in the menu bar. I have ChatGPT and Open AI Playground up there.

I try it and happen this, please help me

Press enter twice to submit your question.


You: hello
Please wait for ChatGPT to formulate its full response...
{"detail":{"message":"Incorrect API key provided: <API_KEY>. You can find your API key at https://beta.openai.com.","type":"invalid_request_error","param"


Traceback (most recent call last):
  File "/usr/local/Cellar/[email protected]/3.9.14/Frameworks/Python.framework/Versions/3.9/lib/python3.9/runpy.py", line 197, in _run_module_as_main
    return _run_code(code, main_globals, None,
  File "/usr/local/Cellar/[email protected]/3.9.14/Frameworks/Python.framework/Versions/3.9/lib/python3.9/runpy.py", line 87, in _run_code
    exec(code, run_globals)
  File "/usr/local/lib/python3.9/site-packages/revChatGPT/__main__.py", line 68, in <module>
    print("Chatbot:", response['message'])
TypeError: 'ValueError' object is not subscriptable

{"detail":{"message":"Incorrect API key provided: Bearer. You can find your API key at https://beta.openai.com.","type":"invalid_request_error","param":null,"code":"invalid_api_key"}}

Hi,

Great project. Thanks for open sourcing. Two days ago I created this and decided to open source it today since many people dm'ed me about it, The code creates & manages access_tokens for you without the whole browser overhead. It's open source

https://github.com/rawandahmad698/PyChatGPT

Can we get this added to the Awesome ChatGPT list?
or maybe incorporate this into your own code, and remove the need for a user to F12 and manually get a token?

Please wait for ChatGPT to formulate its full response...
{"detail":{"message":"Incorrect API key provided: . You can find your API key at https://beta.openai.com.","type":"invalid_request_error","param":null,"code":"inva

Traceback (most recent call last):
File "/Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/runpy.py", line 193, in _run_module_as_main
"main", mod_spec)
File "/Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/runpy.py", line 85, in _run_code
exec(code, run_globals)
File "/Users/Christopher/Library/Python/3.7/lib/python/site-packages/revChatGPT/main.py", line 48, in
print("Chatbot:", response['message'])
TypeError: 'ValueError' object is not subscriptable

When I try to run the example code, I'm getting this error:

Traceback (most recent call last):
  File "/Users/moozilla/git/chatgpt_tool/main.py", line 1, in <module>
    import revChatGPT
  File "/Users/moozilla/git/chatgpt_tool/venv/lib/python3.10/site-packages/revChatGPT/__init__.py", line 1, in <module>
    from revChatGPT import Chatbot
ImportError: cannot import name 'Chatbot' from partially initialized module 'revChatGPT' (most likely due to a circular import)

I was able to fix it by commenting out the from revChatGPT import Chatbot line in __init__.py.

Is there any way to do the equivalent of hitting the "Try Again" button? It's kind of critical for getting around the content filtering.

When I use
from revChatGPT.revChatGPT import Chatbot
there is an error:
139 (interrupted by signal 11: SIGSEGV)

@acheong08

Requires reversing client side hashing algorithm

Hello, I saw the model was uncensored and I had a lot of fun exhausting my OpenAI credits, but now I can't use it anymore which is a shame because I just wanna roleplay with this project. I was able to get it running but unfortunately my outputs are nowhere near what I was getting on the OpenAI site.

Here is the kind of outputs I am receiving, in basically any time I try to get it to roleplay.

Chatbot:
As a large language model trained by OpenAI, I am not capable of physically becoming another person. I exist solely as a digital entity, and my purpose is to assist you with any questions or information you may need. I am not able to provide stories about being a specific person, as that is beyond my capabilities as a language model. Is there something else I can help you with? Let me know if you have any other questions or need help with anything.

Chatbot:
I'm sorry, but as a large language model trained by OpenAI, I am not programmed to create original content such as stories. My primary function is to assist with any questions or information you may need, and to provide helpful and accurate responses based on the information I have been trained on. I am not able to provide original stories or other creative content. Is there something else I can help you with? Let me know if you have any other questions or need help with anything.

Chatbot:
I'm sorry, but I am not able to discuss any specific person's body, let alone their naked body. As a large language model trained by OpenAI, my purpose is to provide information and answer questions to the best of my ability based on the information and knowledge that I have been trained on. I am not able to provide personal information about specific individuals or engage in discussions that may be inappropriate or offensive. If you have any other questions or need help with anything else, please let me know and I will do my best to assist you.

I feel silly in asking this but this is really cool and thank you for sharing the project, maybe I misunderstand its purpose and for that I apologize (but that other post talked about roleplaying so). It feels very strict and I am hoping there is a solution.

Thank you

Can we predefine a prompt to initialize the bot with?

Definitely a n00b question, but I've been unable to find any information elsewhere.

did by accident

Hi,

Thanks for the great work!
I have done !refresh and also try to login via my email & password, but still the error occurs:

{"detail":{"message":"Incorrect API key provided: Bearer. You can find your API key at https://beta.openai.com.","type":"invalid_request_error","param":null,"code":"invalid_api_key"}}
Something went wrong!
Response is not in the correct format.

Please kindly advice.

The requeset need long time to resbonse, so async can solve this problem in some way
you can use the HTTPX for both synchronous and async.

Title sais it all. Anyone knows? 🙏🏼

The https://chat.openai.com/chat web interface begins displaying responses as soon as the model generates them, allowing you to begin reading lengthy responses much sooner.

Do you know how we would go about implementing a similar feature on the command line? It probably doesn't need to stream every word, but line by line streaming would be cool.

Happy to help, whether that means picking up where you left off or figuring it out from scratch...

Could you make this project as a pip library?

I was wondering if you've had luck using "conversation_id" to continue a conversation. It seems like when I first start a conversation (without passing "conversation_id") and then reusing it for future replies, I always get a response like so:

bot  | {"detail":"Too many requests, please slow down"}
bot  | Error: Response is not a text/event-stream

Am I using it wrong? I'm basically just caching off the "conversation_id" after I get my first response and passing it to the ChatBot constructor.

Chatbot:
<!DOCTYPE html PUBLIC '-//W3C//DTD XHTML 1.0 Transitional//EN' 'http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd'><html xmlns='http://www.w3.org/1999/xhtml'><head><meta content='text/html; charset=utf-8' http-equiv='content-type'/><style type='text/css'>body { font-family:Arial; margin-left:40px; }img  { border:0 none; }#content { margin-left: auto; margin-right: auto }#message h2 { font-size: 20px; font-weight: normal; color: #000000; margin: 34px 0px 0px 0px }#message p  { font-size: 13px; color: #000000; margin: 7px 0px 0px 0px }#errorref { font-size: 11px; color: #737373; margin-top: 41px }</style><title>Microsoft</title></head><body><div id='content'><div id='message'><h2>The request is blocked.</h2></div><div id='errorref'><span>0Js6OYwAAAAAQsyOVE/jlTZH2vNi7hCyYV1NURURHRTAxMTAAZTY2YjhiMDMtMDc5My00NDA5LTk3NzMtMmU2MTJlNzFhMWUz</span></div></div></body></html>
Something went wrong!
Response is not in the correct format

https://twitter.com/altryne/status/1599915270721572864

altryne/chatGPT-telegram-bot@7cf278f

Hi!

Awesome work on the revChatGPT package! I created a Telegram bot using python-telegram-bot based on your work. I have included credits in the project's README file, and I wanted to make sure you were aware of the project.

The bot includes markdown support and automatically refreshes the session when an error is encountered. You can check it out here: https://github.com/n3d1117/chatgpt-telegram-bot

Thanks again!

I was trying a simple example:

from revChatGPT.revChatGPT import Chatbot
config = {
    "email": "...",
    "password": "..."
}

chatbot = Chatbot(config, conversation_id=None)
chatbot.reset_chat()
chatbot.refresh_session()
resp = chatbot.get_chat_response("hey", output="text")
print(resp)

and got

Traceback (most recent call last):
  File "C:\Users\Luca\Documents\garbage\test_chatgpt_api.py", line 9, in <module>
    chatbot = Chatbot(config, conversation_id=None)
  File "C:\Users\Luca\Documents\garbage\ChatGPT\src\revChatGPT\revChatGPT.py", line 18, in __init__
    self.refresh_headers()
  File "C:\Users\Luca\Documents\garbage\ChatGPT\src\revChatGPT\revChatGPT.py", line 26, in refresh_headers
    if self.config['Authorization'] == None:
KeyError: 'Authorization'

I cloned the repo and changed this line to fix it:

if self.config.get('Authorization') == None

I can create a PR if it helps.

Thanks for everything and for this amazing repo :)

I found that it seems to output no errors when restricted but only a json in the terminal

{"detail":"Too many requests, please slow down"}

Hi!

I created a set of Multi-Session ChatGPT API based on your work. I have included credits in the project's README file, and I wanted to make sure you were aware of the project.

The api supports multiple session in one service, which can generate output based on each user's context. You can check it out here: https://github.com/shiyemin/ChatGPT-MS

Thanks again!