Mimina Osawa and Pina Sformklan Estoh from Ladies Vs Butlers!
Pina 1: long hair, grey hair, double bun, twintails, purple eyes, tiara, blue ribbon, pink blouse, short sleeves, puffy sleeves, red skirt, thighhighs
Pina 2: long hair, grey hair, double bun, twintails, purple eyes, tiara, magical girl, fingerless gloves, cape, panties, thigh strap, single thighhigh, leotard, sleeveless
Mimina 1: flat chest, short hair, brown hair, orange eyes, two side up, hair ribbon, brooch, green ribbon, sweater, red skirt
Mimina 2: flat chest, short hair, brown hair, orange eyes, two side up, magical girl, animal hat, sleeveless, gloves, frills, short dress, thighhighs
Description
FAQ
Comments (12)
Not gonna lock the discussion?
Anyway thanks for the characters!
Unlike Digimon or Fate I do not think people cares for this series enough to worry for the lora gallery getting flood with tos violation porn. XD
Just being curious ^^
Are you gonna train more girls from the series?
I have the intention of doing 2-3 more of these two character sets, but I dont know when
Any plans to remake this for Illusttrious?
Well, as a "plan" Yes, I want to redo this, and I want still do drillhair ojousama at the very least, but I don't have that much time nowdays and the things I want to port to illustrious keep piling up.
@neclordx I get not having enough time, just knowing you plan to port them eventually is enough so thank you.
Wait what? You already made an IL version for this???? Either I didn't notice when I commented or you did it in the last day, either way thank you so much!
@Lancelot69lol I did it last day, after you commented it. I don't mind to explain.
After doing the lora I usually delete the datasets, I do this on purpose, consciously, thinking if I have to it for another model in the future I will gather the data again, Because I feel if you use the same data over and over again instead of trying to improve the datasets the new loras for new models feel...stalled. I aim for them to be better, not just being on new models.
But sometimes the datasets survives the purge, so remaking some lora its just picking the file and push "train". This one was done not long ago and was still lying around in my pc, so I just rebaked it for illustrious.
@neclordx Interesting thought process on avoding stagnation but sometimes images online get deleted so you may end up being unable to get as many images. I think I get the idea but personally I'd just keep the data sets and maybe change/add some new images to improve along with trying new tags for training.
You could even compare your old data sets to the new ones and see how much of a difference there is. Either way I appreciate your work ethic to preoduce the best Lora.
@Lancelot69lol Of course, is case-dependant as well, I don't expect by this point to get new Suikoden II material for example, so those sets I dont delete.
@neclordx I'm relieved to hear that lol, nothing worse than losing images forever.
Details
Files
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.