Issue with "extra_body" Parameter in "llm_config" Using Autogen

[saisuryateja1436]

Hello everyone,
I've been working with the llm_config in Autogen by Microsoft and encountered an issue when trying to use the extra_body parameter. I have successfully implemented this parameter using langchain_openai, but when I attempt to replicate the setup with Autogen, it doesn't seem to work as expected.
Here's a simplified version of the code that works with langchain_openai:

 
from langchain_openai import ChatOpenAI
llm = ChatOpenAI(
    base_url="my_URL",
    api_key="EMPTY",
    default_headers={'Authorization': f"Bearer {token}", 'Content-Type': "application/json"},
    http_client=httpx.Client(verify=False),
    model="mistralai/Mistral-7B-Instruct-v0.1",
    extra_body={
        "A": "1",
        "B": "2",
        "C": "3",
        "D": 4,
    },
)

However, when I try to configure the same setup using Autogen, the extra_body parameter doesn't seem to be applied correctly:

config_list = [
    {
        "model": "mistralai/mistral-7b-instruct-v0.1",
        "api_key": "EMPTY",
        "default_headers": {"Authorization": f"Bearer {os.environ['token']}", 'Content-Type': "application/json"},
        "api_type": "openai",
        "base_url": 'MY_URL',
        "http_client": MyHttpClient(proxy=""),
        "extra_body": {
            "A": "1",
            "B": "2",
            "C": "2",
            "D": 4
        }
    },
]
 
llm_config = {
    "cache_seed": 47,
    "temperature": 0,
    "config_list": config_list,
    "timeout": 120,
 
}
 

I would appreciate any insights or suggestions on how to resolve this issue. Is there a specific way to implement extra_body in Autogen that differs from langchain_openai? Any guidance or examples would be greatly appreciated.
Thank you in advance for your help!