Update (9-6-25): I was provided more images for the dataset, so I did a V2.

CivitAI user ElJyu provided me with training data for this style model and asked if I could train it. Here is the result, which I think looks pretty good and does a pretty good job adhering to the source material! I put phantomspiritstyle as the trigger word, so make sure to have this in your prompt!

➼Obligatory self-promotion: consider supporting or commissioning me at https://www.fiverr.com/metalchromex/make-you-a-stable-diffusion-anime-character-model or https://www.patreon.com/MetalChromeX so I can continue making cool crap for you all to enjoy for free!