Update README.md
Browse files
README.md
CHANGED
|
@@ -8,10 +8,10 @@ tags:
|
|
| 8 |
- ai-toolkit
|
| 9 |
widget:
|
| 10 |
- text: A woman character with red hair, playing chess at the park, bomb going off
|
| 11 |
-
in the background
|
| 12 |
output:
|
| 13 |
url: samples/1725458804843__000008050_0.jpg
|
| 14 |
-
- text: a
|
| 15 |
a cafe
|
| 16 |
output:
|
| 17 |
url: samples/1725458846417__000008050_1.jpg
|
|
@@ -19,10 +19,10 @@ widget:
|
|
| 19 |
laser lights, punk rocker FluxToon style
|
| 20 |
output:
|
| 21 |
url: samples/1725458887751__000008050_2.jpg
|
| 22 |
-
- text: hipster with a beard, building a chair, in a wood shop
|
| 23 |
output:
|
| 24 |
url: samples/1725458928911__000008050_3.jpg
|
| 25 |
-
- text: '
|
| 26 |
output:
|
| 27 |
url: samples/1725458969982__000008050_4.jpg
|
| 28 |
base_model: black-forest-labs/FLUX.1-dev
|
|
@@ -39,6 +39,7 @@ Model trained with [AI Toolkit by Ostris](https://github.com/ostris/ai-toolkit)
|
|
| 39 |
## Trigger words
|
| 40 |
|
| 41 |
You should use `FluxToon style` to trigger the image generation.
|
|
|
|
| 42 |
|
| 43 |
## Download model and use it with ComfyUI, AUTOMATIC1111, SD.Next, Invoke AI, etc.
|
| 44 |
|
|
@@ -54,7 +55,7 @@ import torch
|
|
| 54 |
|
| 55 |
pipeline = AutoPipelineForText2Image.from_pretrained('black-forest-labs/FLUX.1-dev', torch_dtype=torch.bfloat16).to('cuda')
|
| 56 |
pipeline.load_lora_weights('Norod78/FluxToon_style_flux_lora', weight_name='FluxToon_style_flux_lora')
|
| 57 |
-
image = pipeline('A woman character with red hair, playing chess at the park, bomb going off in the background
|
| 58 |
image.save("my_image.png")
|
| 59 |
```
|
| 60 |
|
|
|
|
| 8 |
- ai-toolkit
|
| 9 |
widget:
|
| 10 |
- text: A woman character with red hair, playing chess at the park, bomb going off
|
| 11 |
+
in the background FluxToon style
|
| 12 |
output:
|
| 13 |
url: samples/1725458804843__000008050_0.jpg
|
| 14 |
+
- text: a FluxToon style man character holding a coffee cup, in a beanie, sitting at
|
| 15 |
a cafe
|
| 16 |
output:
|
| 17 |
url: samples/1725458846417__000008050_1.jpg
|
|
|
|
| 19 |
laser lights, punk rocker FluxToon style
|
| 20 |
output:
|
| 21 |
url: samples/1725458887751__000008050_2.jpg
|
| 22 |
+
- text: hipster with a beard, building a chair, in a wood shop FluxToon style
|
| 23 |
output:
|
| 24 |
url: samples/1725458928911__000008050_3.jpg
|
| 25 |
+
- text: 'FluxToon style Wonderwoman holding a sign that says ''I LOVE FluxToon!'''
|
| 26 |
output:
|
| 27 |
url: samples/1725458969982__000008050_4.jpg
|
| 28 |
base_model: black-forest-labs/FLUX.1-dev
|
|
|
|
| 39 |
## Trigger words
|
| 40 |
|
| 41 |
You should use `FluxToon style` to trigger the image generation.
|
| 42 |
+
When describing people, adding the word 'character' helps (e.g. 'a woman character', 'a group of characters' and so on)
|
| 43 |
|
| 44 |
## Download model and use it with ComfyUI, AUTOMATIC1111, SD.Next, Invoke AI, etc.
|
| 45 |
|
|
|
|
| 55 |
|
| 56 |
pipeline = AutoPipelineForText2Image.from_pretrained('black-forest-labs/FLUX.1-dev', torch_dtype=torch.bfloat16).to('cuda')
|
| 57 |
pipeline.load_lora_weights('Norod78/FluxToon_style_flux_lora', weight_name='FluxToon_style_flux_lora')
|
| 58 |
+
image = pipeline('A woman character with red hair, playing chess at the park, bomb going off in the background FluxToon style').images[0]
|
| 59 |
image.save("my_image.png")
|
| 60 |
```
|
| 61 |
|