JosefJilek
commited on
Commit
•
fe4ca3d
1
Parent(s):
a81b66d
Update README.md
Browse files
README.md
CHANGED
@@ -42,9 +42,21 @@ v0.8.x \
|
|
42 |
Major improvement on higher resolutions \
|
43 |
Style improvements \
|
44 |
Flexibility and responsivity \
|
45 |
-
Added support for Night Sky YOZORA model
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
46 |
|
47 |
-
## Examples
|
48 |
### AOM
|
49 |
<img src="https://huggingface.co/JosefJilek/loliDiffusion/resolve/main/xyz_grid-0010-1.png"></img>
|
50 |
### Counterfeit
|
|
|
42 |
Major improvement on higher resolutions \
|
43 |
Style improvements \
|
44 |
Flexibility and responsivity \
|
45 |
+
Added support for Night Sky YOZORA model \
|
46 |
+
\
|
47 |
+
v0.9.x \
|
48 |
+
Different approach at merging, you might find v0.8.x versions better
|
49 |
+
Changes at supported models \
|
50 |
+
\
|
51 |
+
v2.1.X EXPERIMENTAL RELEASE \
|
52 |
+
Stable Diffusion 2.1-768 based \
|
53 |
+
Default negative prompt: (low quality, worst quality:1.4), (bad anatomy), extra finger, fewer digits, jpeg artifacts \
|
54 |
+
For positive prompt it's good to include tags: anime, (masterpiece, best quality) alternatively you may achieve positive response with: (exceptional, best aesthetic, new, newest, best quality, masterpiece, extremely detailed, anime, waifu:1.2) \
|
55 |
+
Though it's Loli Diffusion model it's quite general purpose \
|
56 |
+
The ability to generate realistic images as Waifu Diffusion can was intentionally decreased \
|
57 |
+
This model performs better at higher resolutions like 768\*X or 896\*X \
|
58 |
|
59 |
+
## Examples (OUTDATED BUT SHOULD BE REPRESENTATIVE ENOUGH)
|
60 |
### AOM
|
61 |
<img src="https://huggingface.co/JosefJilek/loliDiffusion/resolve/main/xyz_grid-0010-1.png"></img>
|
62 |
### Counterfeit
|