I trained this on the FFT FMVs hoping to recreate that art style. It should be able to recognize the four characters you see in the cutscenes if you prompt their names and features but it's not very consistent. I also tried to tag the chocobo but it didn't work out well, but it did pick up on the very aesthetic candles lol.
I've also uploaded an earlier version that's generally worse but I sometimes mix them both together, it can kind of give an effect closer to what I wanted.
Description
FAQ
Comments (2)
Looks like "Memaxl flat anime style" (as used in one of the example images) is pretty much a requirement to get the full effect when using AutismMix as base.
It was actually kind of hard for me to find images with plain prompts since I usually have several loras loaded lol. Thanks for the advice though, I wasn't really sure what that lora was until now. It was something someone else was using for a prompt I was trying out.



