Confirm this is an issue with the Python library and not an underlying OpenAI API

• This is an issue with the Python library

Describe the bug

We are getting a RemoteProtocolError often after running a streaming chat completion call for a long time. It always throws after a few minutes (Python 3.11 / FastAPI)

Model: o3-mini

Packages:
openai 1.78.1
httpx 0.28.1

To Reproduce

from fastapi import FastAPI, Request
import openai
import httpx

openai.api_key = "YOUR_KEY"

app = FastAPI()

@app.post("/chat-stream")
async def chat_stream(request: Request):
# parse incoming JSON (e.g. {"messages":[...]}):
body = await request.json()
# kick off OpenAI streaming chat:
stream = await openai.ChatCompletion.acreate(
model="o3-mini",
messages=body["messages"],
stream=True,
)
async def event_generator():
try:
async for chunk in stream:
yield chunk.choices[0].delta.get("content", "")
finally:
# cleanup if needed
pass

return httpx.StreamingResponse(event_generator(), media_type="text/plain")

Code snippets

OS

macOS

Python version

Python 3.11.11

Library version

openai v1.78.1