Am I crazy for thinking the new GPT-4 Turbo is extremely worse than regular GPT-4?

I've been using GPT-4 for the past two years and I felt like it was the best possible tool for me. I've never encountered such a well rounded, intelligent, and mature model, even among other LLM's from different companies. <br><br>I was down to try the new GPT-4 Turbo, but it feels like a completely watered down version of GPT-4. It can't do many of the advanced features the original can do, like code interpretation, to theorize or down right lie. I've also noticed that the model have a more "joke-y" behavior, like it's trying to be extremely funny all the time, which is incredibly annoying.<br><br>Am I crazy or am I right? I know that the new model is theorically more intelligent than regular GPT-4, but I've used it for the past couple of days and none of my interactions have felt even close to the quality of GPT-4.<br><br>Does anyone feel the same?