H Lohaus

On my smartphone the new parameter works: 

I tested this for example: ``` from openai import OpenAI client = OpenAI( api_key="asdf", base_url="http://localhost:1337/v1" ) response = client.chat.completions.create( model="gpt-3.5-turbo", extra_body={"provider": "You"}, messages=[{"role": "user", "content": "write a poem about a...

yes, of course. The pydantic error message can maybe be this: https://github.com/xtekky/gpt4free/pull/1617/files#diff-916ba29538dcda9bda09dfaedee4f7bd72bdfde8ddb576320052b7ee46bf0340 I use only there the BaseModel from pydantic. Can you test this changes?

You don't need to change the g4f code. This import should also work: ``` from g4f.Provider.GeminiProGBeCo import GeminiProGBeCo ``` Also you can add them to list in your code: ```...

The proxy support is the native way. ?

I can't create a Vertex Account for me. Sorry Edit: You have to change only the Url maybe: const API_ENDPOINT = "us-central1-aiplatform.googleapis.com"; const URL = `https://${API_ENDPOINT}/v1beta1/projects/${process.env.GOOGLE_KEY}/locations/us-central1/publishers/google/models/gemini-pro:streamGenerateContent`; edit: I added api_base...

I fixed a issue with the courser and ul li lists Maybe it was this

The API usage of some providers are limited. Try a other working providers from the list.

Update to version 0.2.1.9. The last update have fix for this issue.

Maybe your request is incomplete. Model and messages are required. If you forget them, is the error valid.