Support My Work:
Your support helps me create more amazing content and keeps this passion project alive. Every contribution, no matter how small, makes a big difference. Thank you for being awesome!
old text:
Add a ❤️ and follow to support my future works.
Every character in a different version of this LoRa:
Additional info about choised LoRa in "About this version block"
//why i change format:
it's bad if your LoRa's don't sort in automatic1111 lora list by title(title need to be first);
it's bad if your don't see trigger word in name of model(and civitai crop's long names, but for version it have additional characters);
it's not so good to made 20 LoRa's for each title on Civitai, just look like spam.
multicharacter LoRas harder to control->it's bad.
// result is format <title name>_<character trigger word> // model can contain additional TW, like "Pajamas", but this is not necessary//
train models isn't easy and you can support me here:
❤️Boosty Donation link
LoRa trained on this model https://civarchive.com/models/49608/randomizer89merge
Description
it named same as previous because i'm stupid. Different LoRA with some fixes
FAQ
Comments (22)
seems like this hasn't been active in a while but... for the life of me i can't find a lora of Hinatsuru Akina (ai's mom)... any help on this is appreciated.
do it yourself: https://civitai.com/articles/6503/how-to-easily-create-ponypdxl-lora-better-than-80percent-on-site-just-for-500buzz
I don't make dataset for her originally. (yes this lora making guide is for ~mid+ quality lora, not perfect one, but it enought for most purposes (just works little worse))
@randomizer89 i appreciate the information. was trying to avoid doing it myself but since your set was my last hope, seems i have no choice.
thanks again brev, turned out better than expected and only took about an hour as you guide estimated. i used 107 images all screenshots from the anime. prompt manipulation is also quite responsive.
@ryuk190 where you found 107 different frames of this character? lol, hard work :)
@randomizer89 didn't really take too long, screenshots with vlc, browsed them with a photo viewer and then took more screens of the screenshots with snagit with auto rename, auto format, auto save to folder, to expedite the cropping time it would normally require. it did pick up a few anomalies though from portions of other characters in the shots. i'm debating on if i should go back and crop them out with more precision with photoshop.
Would it be possible to get an Illustrious for this??
maybe, if i have free time
@randomizer89 ok, I can commission if needed
@Sweet99 if you commission i do it in few days(top priority) (for some reason i need to do somethings with dataset, can't just upload PDXL one on train, cuz it one of my first PDXL loras and i made it in non good way)
@randomizer89 I understand, I will make the commission through Ko-fi then
@Sweet99 yes, thanks you very much!
@randomizer89 please tell me which commission option to select at your Ko-fi
@Sweet99 ok https://ko-fi.com/c/90545b66bc i create this for it (don't forgot additionals, models with many characters is harder to made)
@randomizer89 would it be easier to split the characters up? So two or more separate loras, I don't mind commissioning multiple loras.
@Sweet99 ~it anyway much more work than on one character lora :) and i think you don't want to pay for each character 5$(problem not in thing you think, it just more images->more tags to check, and for splitted it additionally more image generation/time for create civitai pages), but if want ok i can split.
@randomizer89 honestly I'm ok with paying for each character! I'll count the characters and use your link to pay the money ^^
You can chose how to split up the loras
@Sweet99 ok, if you pay for each i just made each + multichar(bonus one)
@randomizer89 commission has been sent
@Sweet99 thanks!
@Sweet99 i hope it works ~fine
@randomizer89 yes, thank you for your hard work!!! It is working fine for me.