Comment by courseofaction

According to my understanding the blog post, FreeWilly2 performs near or above ChatGPT4 for most test-cases. Is this true?

Am I misunderstanding this? Is this not a big deal?