Update README.md
Browse files
README.md
CHANGED
@@ -7,9 +7,30 @@ tags:
|
|
7 |
- image-generation
|
8 |
- safetensors
|
9 |
widget:
|
10 |
-
- text: myst33d style, furry
|
11 |
base_model: stabilityai/stable-diffusion-xl-base-1.0
|
12 |
---
|
13 |
|
14 |
# Myst33d Style LoRA
|
15 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
7 |
- image-generation
|
8 |
- safetensors
|
9 |
widget:
|
10 |
+
- text: anime, myst33d style, 1boy, furry, looking at viewer, Tokyo city background
|
11 |
base_model: stabilityai/stable-diffusion-xl-base-1.0
|
12 |
---
|
13 |
|
14 |
# Myst33d Style LoRA
|
15 |
+
A barely working LoRA of my vaguely similar art style
|
16 |
+
|
17 |
+
## Prompting
|
18 |
+
Use `myst33d style` to stylize anything, `anime` works well along with `myst33d style`
|
19 |
+
|
20 |
+
## Using with diffusers
|
21 |
+
```
|
22 |
+
import torch
|
23 |
+
from diffusers import StableDiffusionXLPipeline
|
24 |
+
from diffusers.models import AutoencoderKL
|
25 |
+
|
26 |
+
pipe = StableDiffusionXLPipeline.from_pretrained("stabilityai/stable-diffusion-xl-base-1.0", torch_dtype=torch.float16, variant="fp16", use_safetensors=True)
|
27 |
+
pipe.load_lora_weights("myst33d/myst33dLoRA")
|
28 |
+
pipe.to("cuda")
|
29 |
+
|
30 |
+
image = pipe(
|
31 |
+
prompt="myst33d style, 1boy, furry, looking at viewer, city background",
|
32 |
+
num_inference_steps=30,
|
33 |
+
guidance_scale=9,
|
34 |
+
).images[0]
|
35 |
+
# Do whatever with the image here
|
36 |
+
```
|