I just had to know how ChatGPT-4.5 compared to GPT-4o. Here’s what happened when I compared the two models with 7 prompts.