This seems more like a test of how one specific form of prompt writing works on several models.
This prompt seems too long for some models, such as flux where they dont recommend more than 22 tokens, and some models, like the ACTUAL most customizable model, Stable Diffusion, works better with...