Type | |
Stats | 772 |
Reviews | (108) |
Published | Jul 7, 2023 |
Base Model | |
Trigger Words | Kimiko Kimiko outfit fox ears face scar facial mark gold symbol on forehead furry female furry fangs pink eyes +1 more |
Hash | AutoV2 CC8048A66E |
Hey, This is my first LoRA. If you'd like to buy me a coffee, I'd very much appreciate it!
When using this LoRA I find it best to set the weight to around 0.7 give or take.
A resolution of 1280 x 768 or 1280 x 1024 is what I tend to use for good results. I just swap them when generating portrait or landscape.
Checkpoint's:
I like to use NeatNess Fluffy Fur Mix (I prefer the Infinity version)
You can also try Swoonify as well, it also yields great results. I've tried both versions and they both work well.
LoRA's:
I also add a little bit of blindbox/大概是盲盒 when generating usually a weight around 0.4 is nice, I find it adds a little bit of a different style to it.
finally add a little bit of extra detail with Add More Details - Detail Enhancer / Tweaker (细节调整) LoRA around 0.4 or 0.5 is nice.
Textual Inversions:
VAE:
Blessed Vae specifically blessed2.vae.pt.
Trigger words:
Kimiko, Kimiko outfit, pink hair, pink eyes, furry, furry female, fangs, two-tone fur, animal ears, fox ears, white fur, face scar, facial mark, gold symbol on forehead
If you want a rear view pose use the (from behind) trigger word.
All images used to train this LoRA were a resolution of 1024 x 1024.
Note: I didn't make this LoRA with the tails. I will when I get the time to do so.