Oh, I mean the actual AI trace the API call would give, or a log would provide or the "thinking mode".

This article is really helpful, but it's one side of the story.

I'd be surprised if any AI like this is run without detailed logging of any kind. It's kind of important for eval, etc.