with multiprocessing, pickle issue with client.beta.chat.completions.parse output

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

[X] This is an issue with the Python library

Describe the bug

We use multiprocessing to handle the calls. It was working with client.chat.completions.create. Currently, we are trying client.beta.chat.completions.parse with response_format, i got error:

Traceback (most recent call last): File "/anaconda/lib/python3.10/multiprocessing/queues.py", line 244, in _feed obj = _ForkingPickler.dumps(obj) File "/anaconda/lib/python3.10/multiprocessing/reduction.py", line 51, in dumps cls(buf, protocol).dump(obj) _pickle.PicklingError: Can't pickle <class 'openai.types.chat.parsed_chat_completion.ParsedChatCompletion[CalendarEvent]'>: attribute lookup ParsedChatCompletion[CalendarEvent] on openai.types.chat.parsed_chat_completion failed

anyone can help ?

Thanks

To Reproduce

from pydantic import BaseModel
from . import get_token_provider
import openai
import asyncio
import pickle

class CalendarEvent(BaseModel):
    name: str
    date: str
    participants: list[str]

def original_test():

    MODEL = "gpt-4o-2024-08-06"
    ENDPOINT = "gpt-4o-2024-08-06-endpoint"
    token_provider = get_token_provider()

    client = openai.AsyncAzureOpenAI(
    azure_ad_token_provider=token_provider,
    azure_endpoint=ENDPOINT,
    api_version="2024-08-01-preview",
    )

    func = client.beta.chat.completions.parse(
        model=MODEL,
        messages=[
            {"role": "system", "content": "Extract the event information."},
            {"role": "user", "content": "Alice and Bob are going to a science fair on Friday."},
        ],
        response_format=CalendarEvent,
        )

    completion = asyncio.run(func)
    pickle.dumps(completion)

if __name__ == "__main__":
    original_test()

Code snippets

No response

OS

linux

Python version

py3.10

Library version

openai 1.51.0

openai / openai-python