Commit
·
ad61471
1
Parent(s):
6b76265
Update README.md
Browse files
README.md
CHANGED
|
@@ -8,8 +8,6 @@ datasets: embed/EasyNegative
|
|
| 8 |
|
| 9 |
## Descriptions
|
| 10 |
|
| 11 |
-
This model is specifically designed to produce anime-style images and is recommended for use with both LoRA and character LoRA, with the LoRA block weight expansion setting. By adjusting the block weights of the character LoRA and using the MIDD preset, distortions caused by the character LoRA on the overall style can be minimized.
|
| 12 |
-
|
| 13 |
Using this model will result in clean, anatomically-correct images that accurately capture the essence of anime-style art, complete with stunning backgrounds.
|
| 14 |
|
| 15 |
Two models are provided: an 18 MB LoRA model and a full base model that merges LoRA with Anything V4.5. The full model is recommended for training your character model, and is particularly effective for training anime characters using this model.
|
|
|
|
| 8 |
|
| 9 |
## Descriptions
|
| 10 |
|
|
|
|
|
|
|
| 11 |
Using this model will result in clean, anatomically-correct images that accurately capture the essence of anime-style art, complete with stunning backgrounds.
|
| 12 |
|
| 13 |
Two models are provided: an 18 MB LoRA model and a full base model that merges LoRA with Anything V4.5. The full model is recommended for training your character model, and is particularly effective for training anime characters using this model.
|