[English] (google translate)
The lora work well between 0.7 and 0.8, but i recomend to set it to 0.7
This lora was literally the first try xd, i'm going to clean my hard drive today, so I'm uploading it as a backup just in case. I'm still working on improving it, so someday I will upload version 2 improved full hd 4k hdr
Face and body:
V1
1girl, charonlora, grey hair, long hair, blunt bangs, low twintails, grey eyes, dark skin
Default Clothes:
V1
hat, grey dress, high collar, long sleeves, belt
Optional: (use one or multiple)
black gloves, ribbon, hair ribbon, cross, cross print, collar, expressionless
PD: the tags are only for basic prompts of the character, add or remove tags depends on what you want, also some of the Optional tags are untested but exist in the txt file of the dataset
Lora trained with the model Animefull-final-pruned
Images= 19 Resolution= 576x576
Problems
the "high neck" of the dress have problems
sometimes generates images that are too pale
dress too baked
1. The rights to reposted models belong to original creators.
2. Original creators should contact SeaArt.AI staff through official channels to claim their models. We are committed to protecting every creator's rights. Click to Claim
