ai-agents-proxy-configuration-azurechatopenai-ignores-proxy-environment-variable-cb2e754b

THEN Manually create an httpx.Client and httpx.AsyncClient with the desired proxy URL, then pass them as http_client and http_async_client parameters to AzureChatOpenAI. This workaround ensures all outgoing requests use the specified proxy.

import httpx
from langchain_openai import AzureChatOpenAI

_http_client = httpx.Client(proxy="http://your-proxy:port" if proxy else None)
_http_async_client = httpx.AsyncClient(proxy="http://your-proxy:port" if proxy else None)

llm = AzureChatOpenAI(
    azure_endpoint="...",
    openai_api_version="...",
    azure_deployment="...",
    openai_api_key="...",
    model_name="gpt-4-32k",
    http_client=_http_client,
    http_async_client=_http_async_client
)