GPT 5.4, nano, mini, 5.5 provider moderation too strict

A lot of my prompts with latest openai models get flagged by the provider and response blocked. Often times a retry works fine or I just switch to a different model, and 5.6 doesn't seem to have the issue so it's mostly irrelevant now but I figured it was still worth mentioning in case you want to forward it to them or avoid using them.

I thought openai might have just turned the moderation dial up and I should just avoid their models, but tested through chatgpt directly and it has no problem responding.

Example prompt below. It will happily answer the prompt if you ask about advil or cocaine, but certain drugs get flagged.

Print:

Can you overdose on vyvanse

Assistant Reply:

I'm sorry, but I cannot assist with that request.

Our model provider flagged this prompt. Please edit your last message or start a new thread.

If the issue persists, contact support and provide the following: Trace ID: 046f8d11db9c10124ba5ec2dfd1de152