metadata
tags:
- text-to-image
- stable-diffusion
- lora
- diffusers
- template:sd-lora
widget:
- text: A dog in the PaperCutout style
parameters:
negative_prompt: blurry, grainy, unfocused, nsfw, nude, naked
output:
url: images/sdxl-PaperCutouts-Dreambooh_20231117225414_e000005_01_7777.png
- text: A photo of Gal Gadot as wonderwoman PaperCutout style
parameters:
negative_prompt: blurry, grainy, unfocused, nsfw, nude, naked
output:
url: images/sdxl-PaperCutouts-Dreambooh_20231117225404_e000005_00_7777.png
- text: The girl with a pearl earring PaperCutout style
parameters:
negative_prompt: blurry, grainy, unfocused
output:
url: images/sdxl-PaperCutouts-Dreambooh_20231117225424_e000005_02.png
- text: A crazy clown PaperCutout style
parameters:
negative_prompt: blurry, grainy, unfocused
output:
url: images/sdxl-PaperCutouts-Dreambooh_20231117225435_e000005_03_7777.png
- text: Elsa from frozen, Disney style PaperCutout style
parameters:
negative_prompt: blurry, grainy, unfocused
output:
url: images/sdxl-PaperCutouts-Dreambooh_20231117225450_e000005_04_7777.png
- text: Elsa from frozen, Disney style PaperCutout style
parameters:
negative_prompt: blurry, grainy, unfocused
output:
url: images/sdxl-PaperCutouts-Dreambooh_20231117225501_e000005_05_7777.png
base_model: stabilityai/stable-diffusion-xl-base-1.0
instance_prompt: PaperCutout style
license: mit
datasets:
- Norod78/PaperCutoutStyle
Paper Cutout style (SDXL Dreambooth LoRA)
Model description
Just experimenting with Dreambooth LoRA training. Dataset was only 14 images generated with MidJourney. Trigger phrase is "PaperCutout style"
Trigger words
You should use PaperCutout style
to trigger the image generation.
Dataset
Dataset images were generated with Midjourney and are available here
Training config
Used kohya_ss with this config
Download model
Weights for this model are available in Safetensors format.
Download them in the Files & versions tab.