vtuber commited on
Commit
bd75911
·
1 Parent(s): e39feb3

Update README.md

Browse files

Added description + example images

Files changed (1) hide show
  1. README.md +11 -3
README.md CHANGED
@@ -1,3 +1,11 @@
1
- ---
2
- license: unknown
3
- ---
 
 
 
 
 
 
 
 
 
1
+ This lora works best with a lower weight. If the image start getting jagged lines, the weight is too high. It also works best with 2D models.
2
+
3
+ This style works well with other character loras and TIs when the character embedding is at a lower weight.
4
+
5
+ | Example Images | | |
6
+ | ------------- | ------------- | ------------- |
7
+ | <img src="https://imgur.com/3p0gtlc.png" width="200" height ="400"> | <img src="https://imgur.com/fxoze1T.png" width="200" height ="400"> | <img src="https://imgur.com/v3qf8il.png" width="200" height ="400">|
8
+ | <img src="https://i.imgur.com/w4qhDNj.png" width="200" height ="400"> | <img src="https://imgur.com/IwJX4i6.png" width="200" height ="400"> |<img src="https://imgur.com/CqVMOb9.png" width="200" height ="400"> |
9
+
10
+ This was my first style lora, so it could use some improvements, but not sure when I will get around to retraining. It was trained on 48 images.
11
+