Jogn Titor
commited on
Commit
·
e19d324
1
Parent(s):
6656a9b
Update README.md
Browse files
README.md
CHANGED
@@ -1,9 +1,14 @@
|
|
|
|
|
|
|
|
|
|
1 |
☁️ Hello, this is Ether Blu Mix 🔵
|
2 |
|
3 |
This is a simple mix aimed at providing a semi-realistic anime style while retaining the knowledge trained on Danbooru-styled tags. Though its base model, AOM3, is fit for anime styled illustrations it's capable of other aesthetics with the correct style of prompting.
|
4 |
|
5 |
The mix also biases towards females and feminine faces, likely due to their prominence within the models as well as the anime aesthetic leaning towards feminine\androgynous faces.
|
6 |
|
|
|
7 |
▼ Generation Guidelines
|
8 |
|
9 |
Use whatever sampler, steps, cfg you prefer.
|
@@ -50,6 +55,7 @@ More realistic images
|
|
50 |
|
51 |
The sample images were generated using Waifu Diffusion VAE kl-f8-anime2. This VAE is not mandatory. Please use whichever VAE you prefer.
|
52 |
|
|
|
53 |
▼ Model Merging
|
54 |
|
55 |
I have no training or real knowledge on machine learning. I merged the model using the information provided on WarriorMama777's AOM model page. I very arbitrarily used the UBM weights given to merge and create AOM3 and its derivatives and modified the values with trial and error until I was visually happy with it.
|
@@ -98,6 +104,7 @@ Alpha: 1
|
|
98 |
|
99 |
Final Model: EBM-v1
|
100 |
|
|
|
101 |
Models Used:
|
102 |
|
103 |
AOM3.safetensors
|
@@ -116,6 +123,7 @@ pastelmix-better-vae-fp16.safetensors
|
|
116 |
|
117 |
D01A68AE76F97506363F387F5F28BB564AD9E20924844FD5945E600B72D39E79
|
118 |
|
|
|
119 |
Credits:
|
120 |
|
121 |
Thank you to WarriorMama777 for providing AOM and other various mixes as well as detailing your workflow and inspiring me to try mixing my own models.
|
|
|
1 |
+
Also available on CivitAI: https://civitai.com/models/17427/ether-blu-mix-ebm
|
2 |
+
|
3 |
+
|
4 |
+
|
5 |
☁️ Hello, this is Ether Blu Mix 🔵
|
6 |
|
7 |
This is a simple mix aimed at providing a semi-realistic anime style while retaining the knowledge trained on Danbooru-styled tags. Though its base model, AOM3, is fit for anime styled illustrations it's capable of other aesthetics with the correct style of prompting.
|
8 |
|
9 |
The mix also biases towards females and feminine faces, likely due to their prominence within the models as well as the anime aesthetic leaning towards feminine\androgynous faces.
|
10 |
|
11 |
+
|
12 |
▼ Generation Guidelines
|
13 |
|
14 |
Use whatever sampler, steps, cfg you prefer.
|
|
|
55 |
|
56 |
The sample images were generated using Waifu Diffusion VAE kl-f8-anime2. This VAE is not mandatory. Please use whichever VAE you prefer.
|
57 |
|
58 |
+
|
59 |
▼ Model Merging
|
60 |
|
61 |
I have no training or real knowledge on machine learning. I merged the model using the information provided on WarriorMama777's AOM model page. I very arbitrarily used the UBM weights given to merge and create AOM3 and its derivatives and modified the values with trial and error until I was visually happy with it.
|
|
|
104 |
|
105 |
Final Model: EBM-v1
|
106 |
|
107 |
+
|
108 |
Models Used:
|
109 |
|
110 |
AOM3.safetensors
|
|
|
123 |
|
124 |
D01A68AE76F97506363F387F5F28BB564AD9E20924844FD5945E600B72D39E79
|
125 |
|
126 |
+
|
127 |
Credits:
|
128 |
|
129 |
Thank you to WarriorMama777 for providing AOM and other various mixes as well as detailing your workflow and inspiring me to try mixing my own models.
|