license: other
tags:
- text-to-image
- stable-diffusion
- lora
- diffusers
- template:sd-lora
- photorealistic
- style
- caricature
base_model: stabilityai/stable-diffusion-xl-base-1.0
instance_prompt: caricature
widget:
- text: caricature full length portrait Gandalf in an enchanted forest
output:
url: 3275940.jpeg
- text: >-
caricature portrait of abraham lincoln in a tall top hat in a victorian
theater
output:
url: 3275938.jpeg
- text: >-
caricature half-body portrait of jennifer coolidge drinking a cocktail at
a tropical resort. She has a very large head.
output:
url: 3275962.jpeg
- text: >-
caricature portrait of the grinch in a santa hat in front of an
exquisitely decorated christmas tree, rich saturated colors
output:
url: 3276116.jpeg
- text: >-
caricature full body portrait of a boxer dog wearing boxer shorts and
boxing gloves in a boxing ring
output:
url: 3276130.jpeg
- text: >-
caricature full body portrait of {happy|surprised} Katy Perry at
mountain
output:
url: 3276131.jpeg
- text: >-
caricature full length portrait of vivid colors very tall Marilyn Manson
in a dark and gritty abandoned haunted house
output:
url: 3276134.jpeg
- text: >-
caricature full body portrait of Lurch from the Addams family in front of
a grand staircase in a haunted house
output:
url: 3276136.jpeg
- text: >-
caricature full body portrait of a beautiful bride with an enormous head
at in a flower garden surrounded by bouquets of vibrant beautiful flowers
output:
url: 3276143.jpeg
- text: caricature full length portrait of a forest gnome peeking
output:
url: 3276144.jpeg
Caricature XL
This LoRA creates caricature portraits in a photorealistic style using SDXL Base as your... base. The trigger is "caricature" and there are three framings/compositions intentionally trained into it.
closeup
half-body
full body
Try starting your prompt with of of the following (replace the italicized words with your own):
caricature closeup portrait of subject in wearing clothing style in location
caricature half-body portrait of subject in wearing clothing style in location
caricature full body portrait of subject in wearing clothing style in location
Important notes:
If you don't specify a location, the default background is a generic studio backdrop or empty room
The LoRA will try to fill the frame so if you want a tall person use a tall portrait dimension, wide landscape for a wide face, etc.
Add emotion keywords to change the default unhappy expression
A CFG of 3 or 3.5 will help avoid contrast issues. The training images were a little "hot"
Finally I would like to thank Wizard Whitebeard for sharing his (at the time private) Gildenface LoRA with me! It was the inspiration for this LoRA, and the source for my initial training images of horrible and marvelous distorted faces.
Image examples for the model:
caricature portrait of abraham lincoln in a tall top hat in a victorian theater
caricature half-body portrait of jennifer coolidge drinking a cocktail at a tropical resort. She has a very large head.
caricature portrait of the grinch in a santa hat in front of an exquisitely decorated christmas tree, rich saturated colors
caricature full body portrait of a boxer dog wearing boxer shorts and boxing gloves in a boxing ring
caricature full body portrait of {happy|surprised} Katy Perry at mountain
caricature full length portrait of vivid colors very tall Marilyn Manson in a dark and gritty abandoned haunted house
caricature full body portrait of Lurch from the Addams family in front of a grand staircase in a haunted house
caricature full body portrait of a beautiful bride with an enormous head at in a flower garden surrounded by bouquets of vibrant beautiful flowers
caricature full length portrait of a forest gnome peeking