Skip to content
Open
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
46 changes: 46 additions & 0 deletions examples/dynamic_proxy_forwarding.py
Original file line number Diff line number Diff line change
@@ -0,0 +1,46 @@
"""
Example: forwarding OpenAI requests through an intermediary proxy.

Machine A sends requests to Machine B, and Machine B forwards to the OpenAI service while preserving streaming responses.
"""

import os
import httpx
from openai import OpenAI, AsyncOpenAI

# Base URL for your proxy server (Machine B)
proxy_base_url = os.getenv("PROXY_BASE_URL", "http://my.proxy.host:8080/v1")


def run_sync_example() -> None:
"""Synchronously send a chat completion request via the proxy and stream the response."""
client = OpenAI(
base_url=proxy_base_url,
http_client=httpx.Client(proxies=proxy_base_url),
)
with client.chat.completions.stream(
model="gpt-4o",
messages=[{"role": "user", "content": "Hello!"}],
) as stream:
for event in stream:
print(event.content or "", end="", flush=True)


async def run_async_example() -> None:
"""Asynchronously send a chat completion request via the proxy and stream the response."""
async_client = AsyncOpenAI(
base_url=proxy_base_url,
http_client=httpx.AsyncClient(proxies=proxy_base_url),
)
async with async_client.chat.completions.stream(
model="gpt-4o",
messages=[{"role": "user", "content": "Hi!"}],
) as stream:
async for event in stream:
print(event.content or "", end="", flush=True)


if __name__ == "__main__":
# For demonstration purposes, run the synchronous example when executed directly.
run_sync_example()
# To run the asynchronous example, use: asyncio.run(run_async_example())