Everywhere I look, I see people making screencap style checkpoints and LORAs using screencaps from actual animes. I can't help but think that's both tiresome and time-consuming, and am surprised not many people seem to realize there are numerous artists such as Hara3131 out there that draw relatively the same way. Therefore, my approach to this kind of style is to train on those artists' styles instead, afterward providing them here for people to use.
Due to this kind of style being similar universally, keep in mind with some styles there may be very little difference.
Updates to this post will be made periodically. Remember to check the About This Version details for specifics on every LORA release.
Description
Trained on the primarily Black Clover/Seven Deadly Sins-esque style of MandioArt. Artist does mostly NSFW work. Generally combines well with other LORAs.
Issues/Quirks:
Some generations output small graphical errors/artefacts, as if some spots got more fried compared to the rest. Cause undetermined.
If the style isn't coming through, adding "flat color, anime coloring" to the prompt potentially solves this. Naturally, screencap-style checkpoints or checkpoints capable of similar styles allow it to shine the most.
Maintains overall style even if not in screencap form.
Recommended Weight(s): 0.6-1
Trigger token is Mandio, but doesn't work, so ignore.
FAQ
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.