Update README.md
Browse filesAdd sample prompts and outputs, and a description for how to get pixel-perfect images.
README.md
CHANGED
|
@@ -21,7 +21,39 @@ Model trained with [AI Toolkit by Ostris](https://github.com/ostris/ai-toolkit)
|
|
| 21 |
|
| 22 |
## Trigger words
|
| 23 |
|
| 24 |
-
No trigger words defined.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 25 |
|
| 26 |
## Download model and use it with ComfyUI, AUTOMATIC1111, SD.Next, Invoke AI, etc.
|
| 27 |
|
|
|
|
| 21 |
|
| 22 |
## Trigger words
|
| 23 |
|
| 24 |
+
No trigger words defined, but it's recommended to start with "*A portrait of \[subject\] with \[appearance/traits\]*".
|
| 25 |
+
|
| 26 |
+
## Sample outputs
|
| 27 |
+
|
| 28 |
+
Images from left-to-right were generated at 512x512 with:
|
| 29 |
+
|
| 30 |
+
No LoRA, 500 steps, 1000 steps, 1500 steps, 2000 steps (final checkpoint)
|
| 31 |
+
|
| 32 |
+
### Medieval Knight
|
| 33 |
+
|
| 34 |
+
**Prompt**: *A portrait of a medieval knight with iron armor, wearing a full iron helmet with a red plume on the back side of the helm. The knight is facing front right at a 45-degree angle with a neutral pose. The iron armor is clean and reflective, showing a clean polish.*
|
| 35 |
+
|
| 36 |
+
|
| 37 |
+
|
| 38 |
+
### Witch
|
| 39 |
+
|
| 40 |
+
**Prompt**: *A portrait of a witch with long red hair and blue eyes, wearing a dark purple witch hat and robes. The witch is facing left with a neutral expression. The robes are trimmed with white and light purple colors.*
|
| 41 |
+
|
| 42 |
+
|
| 43 |
+
|
| 44 |
+
### Werewolf
|
| 45 |
+
|
| 46 |
+
**Prompt**: *A portrait of a werewolf with light gray fur and yellow eyes, wearing a red scarf and a sword on his back. The werewolf is facing forward with a confident expression.*
|
| 47 |
+
|
| 48 |
+
|
| 49 |
+
|
| 50 |
+
|
| 51 |
+
## How to get pixel-perfect images
|
| 52 |
+
|
| 53 |
+
To get pixel-perfect images, downscale by a factor of 8. So 512x512 images should downscale to 64x64, 1024x1024 to 128x128, and so on.
|
| 54 |
+
|
| 55 |
+
You can generate at higher resolutions like 1024x1024, but you may get worse pixel quality because the LoRA was trained on 512x512 images only.
|
| 56 |
+
|
| 57 |
|
| 58 |
## Download model and use it with ComfyUI, AUTOMATIC1111, SD.Next, Invoke AI, etc.
|
| 59 |
|