First of all, I can no longer use Claude freely. It has been updated only once every five days, which is really frustrating. I just became a Google AI Pro member, and this happened. I feel like I’ve been cheated.
Then there’s Gemini 3 Pro (High). I asked it to handle code work, but it’s like a dumbed-down model. It’s supposed to be Google’s strongest model, yet it often inserts code comparisons in the wrong places, causing a lot of errors in the files. I find this very amusing.
Have you ever encountered such a thing, too?