Update README.md

[skullcrushercmd]

your script for getting liaobots to work is outdated and ass

[skullcrushercmd]

unfinished but usable code in js, will need to convert it into python anyways so you can make whatever changes.

every ip gets 4-5 authcodes which is ab 8k tokens, a cookie is needed and doesnt expire to get past bot auth and you can figure out the rest from the code https://github.com/any-VM/anyVM-bot/blob/main/gpt.js

[hlohaus]

In my opinion, Liaobots is performing adequately and effectively. I do not perceive any immediate areas that require modification or improvement.

[skullcrushercmd]

In my opinion, Liaobots is performing adequately and effectively. I do not perceive any immediate areas that require modification or improvement.

It doesn't check if it can get the auth codes and it only gets a single authcode, youre able to at least get 5 of them but even then the script can at least throw a meaningful error when you can't get anymore authcodes/ ran out of credits with a suggestion to use a VPN, also implementing a proxy connection could allow you to gather more authcodes, it's shouldn't be a matter if it works rather does it work well. If you really think that way no wonder so many of the providers doesn't work among other things.

[hlohaus]

Hey, so I've been using it sometimes, and it's working perfectly! I use it because of the free 5 GPT-4 messages. Maybe we should add a note to the readme?

Edit: We could reset the cookie jar after 5 requests, but it won't work because of the IP check.

[skullcrushercmd]

Hey, so I've been using it sometimes, and it's working perfectly! I use it because of the free 5 GPT-4 messages. Maybe we should add a note to the readme?

Edit: We could reset the cookie jar after 5 requests, but it won't work because of the IP check.

are you mixing up you.com with liaobots? they dont do free messages they just give authcodes (equivalent to api keys) with .1 credits on it, also the cookie is only used for bot verification and havent had one expire on me (if you dont have a cookie it wont let you make a requets to their api, instead just sending you a bot page to get verified/ get a cookie)

[skullcrushercmd]

also if you have discord id rather talk on there and maybe pick your brain on how g4f is structured overall because im making a js adaptation and it would help asking someone who works on it. my username on discord is the same on gh

[skullcrushercmd]

also the script for you.com seems to be broken, specifically getting the telemetry id or smth

Using You provider and gpt-4 model No .har file found Getting telemetry_id for you.com with nodriver ERROR:root: Traceback (most recent call last): File "C:\Users\tooni\Documents\github\gpt4free\g4f\gui\server\api.py", line 174, in _create_response_stream for chunk in ChatCompletion.create(**kwargs): File "C:\Users\tooni\Documents\github\gpt4free\g4f\providers\base_provider.py", line 216, in create_completion yield loop.run_until_complete(await_callback(gen.anext)) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "C:\Python312\Lib\asyncio\base_events.py", line 685, in run_until_complete return future.result() ^^^^^^^^^^^^^^^ File "C:\Users\tooni\Documents\github\gpt4free\g4f\providers\base_provider.py", line 45, in await_callback return await callback() ^^^^^^^^^^^^^^^^ File "C:\Users\tooni\Documents\github\gpt4free\g4f\Provider\You.py", line 79, in create_async_generator cookies = await cls.get_cookies(session) if chat_mode != "default" else None ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "C:\Users\tooni\Documents\github\gpt4free\g4f\Provider\You.py", line 159, in get_cookies cls._cookies = await cls.create_cookies(client) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "C:\Users\tooni\Documents\github\gpt4free\g4f\Provider\You.py", line 186, in create_cookies cls._telemetry_ids = await get_telemetry_ids() ^^^^^^^^^^^^^^^^^^^^^^^^^ File "C:\Users\tooni\Documents\github\gpt4free\g4f\Provider\you\har_file.py", line 90, in get_telemetry_ids browser = await start() ^^^^^^^^^^^^^ File "C:\Python312\Lib\site-packages\nodriver\core\util.py", line 74, in start return await Browser.create(config) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "C:\Python312\Lib\site-packages\nodriver\core\browser.py", line 87, in create await instance.start() File "C:\Python312\Lib\site-packages\nodriver\core\browser.py", line 312, in start await asyncio.create_subprocess_exec( File "C:\Python312\Lib\asyncio\subprocess.py", line 224, in create_subprocess_exec transport, protocol = await loop.subprocess_exec( ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "C:\Python312\Lib\asyncio\base_events.py", line 1742, in subprocess_exec transport = await self._make_subprocess_transport( ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "C:\Python312\Lib\asyncio\base_events.py", line 523, in _make_subprocess_transport raise NotImplementedError NotImplementedError 127.0.0.1 - - [07/May/2024 17:52:10] "POST /backend-api/v2/conversation HTTP/1.1" 200 - INFO:werkzeug:127.0.0.1 - - [07/May/2024 17:52:10] "POST /backend-api/v2/conversation HTTP/1.1" 200 - Closing port 8080 Closing port 8080

im on python 12, tried 11 aswell and same issue so idk

[hlohaus]

Hey, it need either a nodriver or a .har file for You.com. The error messages are coming from nodriver, and it needs the original event loop. Did you change the event loop or are you using duckduckgo-search?

You can write me on discord: hlohaus789

[github-actions[bot]]

Bumping this pull request because it has been open for 7 days with no activity. Closing automatically in 7 days unless it becomes active again.

[hlohaus]

I updated the Liabots provider. Two types of authentication are now used.