Moderation endpoint is missing a timeout/request_timeout parameter

[danieljannai21]

Describe the feature or improvement you're requesting

Just like any other endpoint, a timeout is very much needed for the moderation endpoint.

Additional context

No response

[jhoseph88]

@danieljannai21 Thanks for catching this! I also ran into this and went ahead and submitted #475 to address it.

[RobertCraigie]

This is fixed in the v1 SDK!

A with_options method is provided that allows you to customise client level defaults on a per-request basis

import httpx
from openai import OpenAI

client = OpenAI()

completion = client.with_options(timeout=httpx.Timeout(connect=2)).chat.completions.create(
    model="gpt-4",
    messages=[
        {
            "role": "user",
            "content": "Say hi!",
        }
    ],
)

[jhoseph88]

@RobertCraigie That's fantastic! Thank you so much for the update. 🚀