Pro Thinking Doesn't Use Pro Model?
Hi all,
I am a bit confused at the moment and was hoping for some clarification.
I was under the impression that the 'Pro thinking' status on a question given to GPT Pro 5.5 meant that the system was processing the reply using the GPT 5.5 Pro model, but after chatting with OpenAI support they said that 'Pro Thinking' actually means Pro effort applied to the GPT 5.5 Thinking model, not the actual GPT 5.5 Pro model itself.
I tried twice on two different AI support windows and have escalated to email support which I am still waiting to hearing back from.
Did I run into hallucinating AI support assistants, or is this actually how it works? It seems insanely misleading, especially considering that the chat windows all say 'Used GPT-5.5 Pro' when hovering over the retry button and have icons labeled as Pro feedback.
Edit: One more thing that I should have mentioned. This started because I was using app calls to Github while using the Pro model. According to OpenAI's doc's this is not support by 5.5 Pro, which is fine, but if you make a tool call while using GPT 5.5 Pro the switch to a thinking or instant model is completely hidden. It still says Pro Thinking, still says Used GPT-5.5 Pro', and still has a button for Pro Feedback despite having definitively used a thinking model for the entire reply.