added Superposition workflow for ComfyUI
Browse files
README.md
CHANGED
@@ -3,6 +3,32 @@ license: creativeml-openrail-m
|
|
3 |
base_model: runwayml/stable-diffusion-v1-5
|
4 |
---
|
5 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
6 |
# Weird Blerbie's Storytime
|
7 |
|
8 |
This model is my Storytime model, but with the Weird Blerbie mod (I multiplied all of the UNET weights by 0.95)
|
|
|
3 |
base_model: runwayml/stable-diffusion-v1-5
|
4 |
---
|
5 |
|
6 |
+
# Update: Introducing Superposition
|
7 |
+
|
8 |
+
WeirdBlerbie was feeling a bit left out when I came up with the Concept Smasher (see the Storytime model card).
|
9 |
+
|
10 |
+
So, we can't have that, can we? Introducing the Superposition workflow for ComfyUI:
|
11 |
+
|
12 |
+
![](Superposition/Workflow.PNG)
|
13 |
+
|
14 |
+
|
15 |
+
Sometimes you need five prompts, all at once. Shown above we're playing with the words "space" and "time". That produces images such as these, with variation on the CFG amount:
|
16 |
+
|
17 |
+
![](Superposition/Space-Time_00001_.png)
|
18 |
+
|
19 |
+
![](Superposition/Space-Time_00003_.png)
|
20 |
+
|
21 |
+
![](Superposition/Space-Time_00017_.png)
|
22 |
+
|
23 |
+
![](Superposition/Space-Time_00048_.png)
|
24 |
+
|
25 |
+
Any model may be used as input. WARNING: IF A MODEL HAS NSFW TRAINING DATA YOU CAN GET NSFW OUTPUTS AT ANY TIME. It is important to understand that, even
|
26 |
+
with this much conditioning, that can happen depending on the model you're using.
|
27 |
+
|
28 |
+
The Superposition workflow json is available here:
|
29 |
+
|
30 |
+
![Superposition workflow json](Superposition.json)
|
31 |
+
|
32 |
# Weird Blerbie's Storytime
|
33 |
|
34 |
This model is my Storytime model, but with the Weird Blerbie mod (I multiplied all of the UNET weights by 0.95)
|