Open iamsky13 opened 6 months ago
Same. I am using async variant inside discord bot. It was working fine yesterday.
This issue is that chatgpt has changed how they handle messages. they are using wss instead. someone can please update this library so it can handle wss web scokets. my python skills are limited i would have done it.
Same promlem
This issue is that chatgpt has changed how they handle messages. they are using wss instead. someone can please update this library so it can handle wss web scokets. my python skills are limited i would have done it.
As you said, ChatGPT changed their backend infrastructure, now they are using websockets. But it seems like this change has only been made in some regions because for some people the re_gpt is still working (including me). And after monitoring ChatGPT's network traffic in my browser, no ebsocket connection was made, it's using the method it was using before. Thus, I can't debug this until the new changes have been applied in my region too.
still not working
not working for me eather
FYI
After I saw the latest PR merged ( #33 ), I tried pulling the source code and installing using setup install
but still:
Traceback (most recent call last):
File "C:\Users\Admin\PycharmProjects\playground\main.py", line 27, in <module>
asyncio.run(main())
File "C:\Users\Admin\AppData\Local\Programs\Python\Python312\Lib\asyncio\runners.py", line 194, in run
return runner.run(main)
^^^^^^^^^^^^^^^^
File "C:\Users\Admin\AppData\Local\Programs\Python\Python312\Lib\asyncio\runners.py", line 118, in run
return self._loop.run_until_complete(task)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\Admin\AppData\Local\Programs\Python\Python312\Lib\asyncio\base_events.py", line 664, in run_until_complete
return future.result()
^^^^^^^^^^^^^^^
File "C:\Users\Admin\PycharmProjects\playground\main.py", line 22, in main
async for message in conversation.chat(prompt):
File "C:\Users\Admin\PycharmProjects\playground\.venv\Lib\site-packages\re_gpt-3.0.2-py3.12.egg\re_gpt\async_chatgpt.py", line 142, in chat
raise UnexpectedResponseError(error, server_response)
re_gpt.errors.UnexpectedResponseError: An unexpected error occurred. Error message: 'NoneType' object is not subscriptable.
This is what the server returned: {"detail":"There was a problem with your chat request. Please use a different browser or try again later."}.
FYI
After I saw the latest PR merged ( #33 ), I tried pulling the source code and installing using
setup install
but still:Traceback (most recent call last): File "C:\Users\Admin\PycharmProjects\playground\main.py", line 27, in <module> asyncio.run(main()) File "C:\Users\Admin\AppData\Local\Programs\Python\Python312\Lib\asyncio\runners.py", line 194, in run return runner.run(main) ^^^^^^^^^^^^^^^^ File "C:\Users\Admin\AppData\Local\Programs\Python\Python312\Lib\asyncio\runners.py", line 118, in run return self._loop.run_until_complete(task) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "C:\Users\Admin\AppData\Local\Programs\Python\Python312\Lib\asyncio\base_events.py", line 664, in run_until_complete return future.result() ^^^^^^^^^^^^^^^ File "C:\Users\Admin\PycharmProjects\playground\main.py", line 22, in main async for message in conversation.chat(prompt): File "C:\Users\Admin\PycharmProjects\playground\.venv\Lib\site-packages\re_gpt-3.0.2-py3.12.egg\re_gpt\async_chatgpt.py", line 142, in chat raise UnexpectedResponseError(error, server_response) re_gpt.errors.UnexpectedResponseError: An unexpected error occurred. Error message: 'NoneType' object is not subscriptable. This is what the server returned: {"detail":"There was a problem with your chat request. Please use a different browser or try again later."}.
PR alreay pushed, wait approved, you can use it at https://github.com/Barrierml/reverse-engineered-chatgpt
still same problem
still same problem
Can you provide a return value for this API use your chatgpt account? https://chat.openai.com/backend-api/sentinel/chat-requirements
still same problem
Can you provide a return value for this API use your chatgpt account? https://chat.openai.com/backend-api/sentinel/chat-requirements
detail" : "Method Not Allowed"
still same problem
Can you provide a return value for this API use your chatgpt account? https://chat.openai.com/backend-api/sentinel/chat-requirements
using get or post?
still same problem
Can you provide a return value for this API use your chatgpt account? https://chat.openai.com/backend-api/sentinel/chat-requirements
using get or post?
get
I found the request in network tab when loading ChatGPT:
{ "persona": "chatgpt-freeaccount", "arkose": { "required": false, "dx": null }, "turnstile": { "required": false }, "token": "MY TOKEN" }
I use pip install -I curl_cffi==0.6.0b7 and it runs ok and if still error try to fix sync_chatgpt.py
I started playing with basic examples and failed to run it