Name is a bit of a misnomer but not sure what else to call it. The model was originally trained on 160 images of various characters from both megaman zero and megaman zx from various artists, the majority of which is actually by Toru Nakayama. Then I merged in some other models that were trained more on specific artists who did some work that included such characters but other stuff as well.

Overall the model is very hit or miss depending on the model it's used with I've personally seen what I consider to be the best results on WinterMoonMix but most models that carry more of a PastelMix style also work well and generally work better with higher weights than models with other anime or realistic styles.

The listed triggers aren't actually necessary but are recognized to some extent by the model.