Golly, Grok 3’s feedback on my latest book work is *significantly* more insightful and useful than ChatGPT 4o’s. But then, 4o isn’t the most advanced model that OpenAI offers.

I tried o3-mini-high, and, as is typically the case, the feedback was much shorter and less detailed. Then I tried o1 and got a better result—more useful detail than either 4o or o3-mini-high—but still noticeably worse than Grok 3.

It bothers me that this org that actively throttles me has made such a good AI. I *really* don’t want to give them money…