LoCon of Symboli Rudolf (umamusume) Please install extension on WebUI.
All of my works is free. If you'd like to support me, feel free to buy me a coffee. 👍
Here are two version of LoCon.
One is resized version using sv_fro method. If you interested in, visit here.
First one used sv_fro method and param=0.99 with kohya_ss resize_lora.py
I think it is good trade-off with file size and quality.
The file size has been reduced from 37.2MB to 23.6MB (It saves 40%!).
The other is original version of LoCon. No resizing.
All preview image settings are same, only LoCon differs.
MAYBE NOT COMPATIBLE WITH AOM3A2, AOM3A3 model.
Recommend options
LoCon weight around 0.6
Trigger words
base
symboli rudolf \(umamusume\)
Racing competition clothes :
red cape, epaulettes, aiguillette, medal, green jacket, long sleeves, white gloves, white ascot, buttons, double-breasted, belt, green skirt, frilled skirt, dress, zettai ryouiki, gold trim, black thighhighs, black footwear
'Hanzo in OW' costume :
official alternate costume, hadanugi dousa, japanese clothes, chest sarashi, open kimono, detached sleeves, single sleeve, ponytail, single bare shoulder, long sleeves, obi, sash, tabi, sandals
!WARNING! 'single bare shoulder' tag maybe impact on nsfw images.
(optional) autumn leaves, autumn
Casual clothes :
white pants, handbag, jewelry, green shirt, casual, long sleeves, sweater, glasses, shoulder bag, belt, adjusting eyewear, collarbone
(optional) adjusting eyewear
Tracen school uniform :
tracen school uniform, purple shirt, pleated skirt, puffy short sleeves, white skirt, puffy sleeves, summer uniform, frilled skirt, sailor collar, sailor shirt, miniskirt, frills, white thighhighs
Use animefull based model
Settings
DPM++ Series(SDE Karras, 2M Karras, etc.)
about 20 steps, CFG scale 3.5~6.5
CLIP skip = 1
Use Hires.fix to get higher quality image
Up-scaler : Latent series
Denoising strength 0.50~0.65
LoRA training info
trained on sd-scripts by kohya_ss. Thank you a lot!
based on Animefull-pruned
148 images of symboli rudolf
4x low quality 52 images
8x medium quality 51 images
16x high quality 45 images
=> 1 epoch = 1336 images
replaced character feature tags with symboli rudolf \(umamusume\)
horse girl, horse tail, brown hair, red eyes, etc.
resolution : 768x768
rank=alpha=16
conv_rank=conv_alpha=16
batch size = 4
other training settings is included in training_data.zip
This LoRA may be compatible with animefull-based models (ex. AbyssOrangeMix Series...)
All uploaded images are generated by AOM3A1
Thank you KohakuBlueleaf for introducing LoCon.
And, special thanks to my best friend. DH.