selecting ChatGPT model and regenerate?

[d-vyd]

Hi. This code works great so far. I have two questions:

(1) I'd like to choose ChatGPT4 sometimes and ChatGPT3.5 other times. When I manually open a new chat with a ChatGPT Plus subscription, I see a model selection dropdown at the top of the page. How can I access that through the object returned by ChatGPT_Client(args.username, args.password) and make the appropriate selection? I'm not proficient in HTML but it doesn't look like a regular dropdown menu to me. Here is the button that allows the user to see the model options:

<button class="relative flex cursor-pointer flex-col rounded-md bg-white py-2 pl-3 pr-10 text-left dark:bg-gray-800 sm:text-sm w-full border border-black/10 focus:border-green-600 focus:outline-none focus:ring-1 focus:ring-green-600 dark:border-white/20" id="headlessui-listbox-button-:r1r:" type="button" aria-haspopup="true" aria-expanded="true" data-headlessui-state="open" aria-labelledby="headlessui-listbox-label-:r1s: headlessui-listbox-button-:r1r:" aria-controls="headlessui-listbox-options-:r2c:"><label class="block text-xs text-gray-700 dark:text-gray-500" id="headlessui-listbox-label-:r1s:" data-headlessui-state="open">Model</label><span class="inline-flex w-full truncate"><span class="flex h-6 items-center gap-1 truncate">GPT-4</span></span><span class="pointer-events-none absolute inset-y-0 right-0 flex items-center pr-2"><svg stroke="currentColor" fill="none" stroke-width="2" viewBox="0 0 24 24" stroke-linecap="round" stroke-linejoin="round" class="h-4 w-4 text-gray-400" height="1em" width="1em" xmlns="http://www.w3.org/2000/svg"><polyline points="6 9 12 15 18 9"></polyline></svg></span></button>

The options are "ChatGPT-3.5", the default, and "ChatGPT-4", which must be selected if desired. Here is the code for that option:

<li class="group relative flex h-[42px] cursor-pointer select-none items-center overflow-hidden border-b border-black/10 pl-3 pr-9 last:border-0 dark:border-white/20 text-gray-900" id="headlessui-listbox-option-:r2h:" role="option" tabindex="-1" aria-selected="false" data-headlessui-state=""><span class="flex items-center gap-1.5 truncate"><span class="flex h-6 items-center gap-1 text-gray-800 dark:text-gray-100">GPT-4</span></span></li>

(2) There are times when I'd prefer to hit the "regenerate" button rather than resubmitting the same question since doing so does not, I believe, contribute to the ChatGPT4 usage limits. How do I access this button?

[ugorsahin]

Hello,

The regenerate function is an important feature, you can find regenerate_response() function in the following commit ecfcc17.

I will update here once I have a solution for model selection, I may need your help to verify the functionality.

[d-vyd]

Wonderful! I'd be happy to test this functionality.

[d-vyd]

Hi. I just wanted to inform you that ChatGPT has changed the model selection widget. Let me know if you need any HTML info about it.

[ugorsahin]

Can you please share a snapshot and html code of this new widget, I can take a look at that now.

[d-vyd]

Let me know if this helps you:

snapshot: https://pasteboard.co/GrfZGp8fBC7e.png
code: https://pastebin.com/yx6q8tzt

[ugorsahin]

Can you please switch to new branch d1a8db3 and check if that works?

[d-vyd]

Thanks! I will check this later today or early tomorrow.

[d-vyd]

Hi. I'm using python 3.10.0. I downloaded and uncompressed the code ZIP. I changed the relative "helpers" import into an absolute import by removing the leading period. I opened a terminal, cd'ed into the directory, and typed python .\chatgpt_client.py. This worked in the "main" branch. In the new branch, I received the error:

Traceback (most recent call last):
  File "C:\Users\david\Downloads\ChatGPT_Automation-9-selecting-chatgpt-model-and-regenerate\ChatGPT_Automation-9-selecting-chatgpt-model-and-regenerate\chatgpt_automation\chatgpt_client.py", line 304, in <module>
    chatgpt = ChatGPT_Client(args.username, args.password)
  File "C:\Users\david\Downloads\ChatGPT_Automation-9-selecting-chatgpt-model-and-regenerate\ChatGPT_Automation-9-selecting-chatgpt-model-and-regenerate\chatgpt_automation\chatgpt_client.py", line 64, in __init__
    version_main=detect_chrome_version()
  File "C:\Users\david\Downloads\ChatGPT_Automation-9-selecting-chatgpt-model-and-regenerate\ChatGPT_Automation-9-selecting-chatgpt-model-and-regenerate\chatgpt_automation\helpers.py", line 10, in detect_chrome_version
    out = subprocess.check_output(['google-chrome', '--version'])
  File "C:\Program Files\Python310\lib\subprocess.py", line 420, in check_output
    return run(*popenargs, stdout=PIPE, timeout=timeout, check=True,
  File "C:\Program Files\Python310\lib\subprocess.py", line 501, in run
    with Popen(*popenargs, **kwargs) as process:
  File "C:\Program Files\Python310\lib\subprocess.py", line 966, in __init__
    self._execute_child(args, executable, preexec_fn, close_fds,
  File "C:\Program Files\Python310\lib\subprocess.py", line 1435, in _execute_child
    hp, ht, pid, tid = _winapi.CreateProcess(executable, args,
FileNotFoundError: [WinError 2] The system cannot find the file specified

[ugorsahin]

• There is an error in version detection because the code is written for linux, I will provide the detection for windows too. Thanks for letting me know about the bug.
• You don't need to change the relative error if you install it. Have you tried using it without installing?
• For now, I am sending a hotfix for version detection ffd2b01, proceed to model switch after instantiating chathead with command ChatGPT_Client.switch_model('GPT-4'). That should work

[d-vyd]

I was pulled onto another project. I apologize for disappearing. I'll be able to report on your recent updates as soon as I get undetected_chromedriver working correctly again on my computer. I've posted my issue here ultrafunkamsterdam/undetected-chromedriver#1563