Welcome to gpt4free Discussions!

[xtekky]

👋 Welcome!

We’re using Discussions as a place to connect with other members of our community. We hope that you:

• Ask questions you’re wondering about.
• Share ideas.
• Engage with other community members.
• Welcome others and are open-minded. Remember that this is a community we
  build together 💪.

To get started, comment below with an introduction of yourself and tell us about what you do with this community.

[jingtianhua]

Hi xtekky,

Do you know the cause of this problem?

[Macx34]

No

[d3vilh]

hehe, you got "Gource visualisation" video on YouTube.
I looked for live video demo and found it.
Keep it going.

[mr-cheffy]

I really love this project! I'm curious what the OpenAI team thinks about this though. Wouldn't this bring some legal issues?

[wh1t3h47]

How does this switch work exactly?

[Ron-Caster]

ChatGPT has limited knowledge i.e; only upto 2021. Allowing the web access in gpt4free make ChatGPT access latest information i.e; upto 2023 and now that is on the internet.

[TheGreatApollyon]

Is there a way to enable this is non-web version?

[orikiel101]

Is there anyone that can create a web based model that works like the default chatgpt3.5 o4.0 smartest non reasoning no fluffs no limits no censor model?

[hlohaus]

I recommend utilizing the free g4f/x.ai provider. I have allocated €10 for Grok 4, which offers rapid reasoning capabilities and an uncensored experience. @orikiel101

[Jdouglas8585]

This is working great so far. I love the interface

[SourabhGPatil]

Check out the Installation process from the repo's Readme file - https://github.com/xtekky/gpt4free#install-

P.S. You can also access this project from this link - https://chat.g4f.ai/chat/

[WhatATechGuy]

I am kinda in the same boat as ManolosGolos. I have completed through step 6, but cannot figure out how to either complete the setup or what I am supposed to do next to interface ChatGPT4. Could someone enlighten me or help me to complete and use gpt4free? -Thanks

[WhatATechGuy]

I put python test.py in Terminal and this is the response:

g4f.Providers.Ails supports: (model: str, messages: list, temperature: float, stream: bool)
Using g4f.Provider.Providers.Forefront provider
Hello
!
How
can
I
assist
you
today
?

Using g4f.Provider.Providers.Bing provider
Traceback (most recent call last):
File "C:\Users\Preston\gpt4free\test.py", line 16, in
response = g4f.ChatCompletion.create(model=g4f.Model.gpt_4, messages=[
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\Preston\gpt4free\g4f_init_.py", line 33, in create
if stream else ''.join(engine._create_completion(model.name, messages, stream, **kwargs)))
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\Preston\gpt4free\g4f\Provider\Providers\Bing.py", line 342, in _create_completion
for token in response:
File "C:\Users\Preston\gpt4free\g4f\Provider\Providers\Bing.py", line 313, in run
next_val = loop.run_until_complete(gen.anext())
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\Preston\AppData\Local\Programs\Python\Python311\Lib\asyncio\base_events.py", line 653, in run_until_complete
return future.result()
^^^^^^^^^^^^^^^
File "C:\Users\Preston\gpt4free\g4f\Provider\Providers\Bing.py", line 286, in stream_generate
raise Exception(
Exception: CaptchaChallenge: User needs to solve CAPTCHA to continue.

Now what I can figure it is trying to interface Bing and get kicked out for not being able to solve a CAPTCHA. Can anyone help?

[ProtDos]

Thats what I call "Open"Ai.

[zjh708]

How does this switch work exactly?

[Brandon-sy]

Amigo como hicistes para poder entrar a gpt4 free GUI por favor ayudame a entrar a esa aplicacion

[EmilMarian47]

Can anyone help me install this? Explain to me in layman terms. Thanks in advance to the creator.

[EmilMarian47]

Thank you for your explanation.

[EmilMarian47]

Installed docker and installed alpine image. Now what?

[djpapzin]

Did you manage to install?

[Tre5co]

I'm in the same boat. I don't know what files to download, or to just download all of them? I can program a gui, I just don't really know how the download works, because I'm assuming the whole program is rather large.

[WhatATechGuy]

What is a/the docker?

[darkzorro79]

[Brandon-sy]

Dude help me get in I don't know how to get into gpt4 free gui help me bro

[SourabhGPatil]

Hey @Brandon-sy, Check out the Installation process from the repo's Readme file - https://github.com/xtekky/gpt4free#install-

P.S. You can also access this project from this link - https://chat.g4f.ai/chat/

[Sinzu007]

How do I go about it

[djpapzin]

did you manage?

[justquick]

Whelp I guess this happened

A developer exploited an API flaw to provide free access to GPT-4

Hope there are more free providers out there. It's now going to turn into a case of whack-a-mole in an arms race between finding access loopholes and the service patching the leaks. The threat of legal action reminds me of this case which never ended up going to trial but non-technical people need to understand where the info is coming from.

https://www.koamnewsnow.com/news/missouri-news/prosecutor-won-t-charge-reporter-who-uncovered-database-flaw/article_615b2860-b04c-5b8d-b325-cfd0bba973df.html

[Cyprien59]

Ok, but could you give me the link to chatgpt4 because I don't have discord.

[djpapzin]

just use Bing Chat

[n1rvanahq]

How to use it in conversation mode? Not just one independent question every time

[djpapzin]

Run this command "streamlit run gui/streamlit_chat_app.py"

[OpenSource-For-Freedom]

Hello. I'm no one, but I am someone within MLops and AI. ChatGPT has assisted in assisting layouts for bash scripting internally at this point. The Linux OS under use can deploy applicable .py Deps much faster. Also, the chat has helped build another AI bot in the works.
I am looking forward to continuing the ML operation moving forward, and remaining secure. And always, Rah, This is the Way.

[rifkiard]

I'm so excited but don't know how to start

[dsglhc]

This is a very good tool, but it has a drawback that if the answer it provides is slightly longer, it will be truncated and even asking again will not display the following content. Is there any solution to this problem?@xtekky

[MasonScarbro]

Can I use this as an API?? if so how

[djfaizp]

Bump

[shorleet]

怎么搞啊，哥

[Saadkhh]

hi guys, im wondering what the difference between these gpt4's is:

[Anthonydt16]

Hi, have people put the gpt4free API into production? If so, does it consume a lot of resources?

[Irkam-1422]

Hey! I am trying to use this tool, but while generating the response I get the following error for every provider: raise RetryProviderError("RetryProvider failed:\n" + "\n".join([
g4f.errors.RetryProviderError: RetryProvider failed:
ChatBase: ClientConnectorCertificateError: Cannot connect to host www.chatbase.co:443 ssl:True [SSLCertVerificationError: (1, '[SSL: CERTIFICATE_VERIFY_FAILED] certificate verify failed: unable to get local issuer certificate (_ssl.c:1000)')]. Could anybody help?

[1234worldcom]

Are u using Vpn then It Is proton vpn with less uptime use another ip

[hemangjoshi37a]

There are so many agents available in the github that use GPT4 and other models, I am thinking of connecting them to g4f instead of the official APIs. How to do this ? please anyone suggest any idea how to implement this .

[1234worldcom]

You Can Make Python Flask server with G4f api and can post Requests by Other codes to get for ex. www.Myserver.com with payload Query= {prompt} and model= {model} and messages= {Messages} And you can check out https://pypi.org/project/Flask/ And https://www.geeksforgeeks.org/flask-tutorial/

[hemangjoshi37a]

this repo does something similar : https://github.com/BerriAI/litellm

[hlohaus]

1. Litellm is cool, but it is restricted.
   g4f is not restricted.

2. You don't need to create a flask server. You can start api server with:

python -m g4f.gui.run

[1234worldcom]

What Is Rate Limit of g4f.Provider.ChatgptAi? Does It reset In 24 Hours and give more Details on it?

[hemangjoshi37a]

actually this repo is something sketchy, i tried and feel this way because you can not depend on it, sometimes it works fine and sometimes it gives errors.

[Woukim]

this repo does something similar : https://github.com/BerriAI/litellm

Did you find a solution?

[ghost]

ok

[ghost]

you got a nice profile bro

[ghost]

Can U please sign in discord i want to talk to u or if u can't then I will contact after some time in WhatsApp @hemangjoshi37a

[hemangjoshi37a]

in plain words i hate the interface of discord and hate discord in plain simple words, if you want to talk to me you are welcome on WhatsApp and telegram on the same number.

[ghost]

ok

[jglorias]

Hi everyone, I keep getting this error, but it doesn't affect running or anything. It just appears when I get a Rate Limit error or something else.

File "/(CENSORED)", line 377, in <module> gendesc = client.chat.completions.create( ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/g4f/client.py", line 163, in create return response if stream else next(response) ^^^^^^^^^^^^^^ File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/g4f/client.py", line 84, in iter_append_model_and_provider for chunk in response: File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/g4f/client.py", line 49, in iter_response for idx, chunk in enumerate(response): File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/g4f/providers/retry_provider.py", line 81, in create_completion raise_exceptions(exceptions) File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/site-packages/g4f/providers/retry_provider.py", line 201, in raise_exceptions raise RetryProviderError("RetryProvider failed:\n" + "\n".join([ g4f.errors.RetryProviderError: RetryProvider failed: Chatgpt4Online: MissingRequirementsError: Install "undetected_chromedriver" and "platformdirs" package

For some reason, there is always a problem with the client.chat.completions.create() function, but as there are no error messages, I am not sure how to fix it. The other error about Missing Requirements, I have already installed the packages undetected_chromedriver and platformdirs, and tried to re-install them again, however the same error happens.

Does anyone have the same error / similar problems? or maybe know how to fix this?
Thank you!

[ghost]

use pip install undetected_chromedriver platformdirs

[1234worldcom]

Hey Can Anyone Join in my project of Free Ai Chat

[ghost]

G4f can improved By adding more providers and check providers and getting best answer by one ai from it

[ghost]

@xtekky Can u Automate Flowgpt with login automation for gpt-4?-turbo?

[andrei-vinnik]

I am trying to use gpt-4-vision to explain an image. This model is asking the image. How to send the image (image_url)?
chat_completion` = client.chat.completions.create( model="gpt-4-vision", messages=[{"role": "user", "content": "Is there a car in the picture?", "image_url": open(path, "rb") }], provider=g4f.Provider.Bing, )
this was not working

this did not work too: "image_url": {"url": f"data:image/jpeg;base64,{encode_image(path)}"}}

[hlohaus]

Simply use image not image_url.

[tracie-lorette]

Please tell me, I used to use g4f and decided to take an example code from git, but now it doesn't work and asks for an api, where can I get it?
Error:
g4f.errors.MissingAuthError: Add a "api_key"

Example code that was running:

from g4f.client import Client
client = Client()
response = client.chat.completions.create(
model="deepseek-v3", # Try "gpt-4o", "deepseek-v3", etc.
messages=[{"role": "user", "content": "Hello"}],
web_search=False
)
print(response.choices[0].message.content)