SDXL LoRA
Collection
A few SDXL-LoRA models I've trained β’ 20 items β’ Updated β’ 8
How to use Norod78/yet-another-sdxl-tattoo-lora with Diffusers:
pip install -U diffusers transformers accelerate
import torch
from diffusers import DiffusionPipeline
# switch to "mps" for apple devices
pipe = DiffusionPipeline.from_pretrained("stabilityai/stable-diffusion-xl-base-1.0", dtype=torch.bfloat16, device_map="cuda")
pipe.load_lora_weights("Norod78/yet-another-sdxl-tattoo-lora")
prompt = "A cthulhu tattoo "
image = pipe(prompt).images[0]A cthulhu tattoo
(CivitAI)
Use the word "tattoo" in your prompts.
I used very simple short prompts. I did not use "Negative" prompt. For some reason "Euler a" with 32 steps (80% SDXL base 20% Refiner) worked best for me.
Trained upon an 1024x1024 resized version of
having the "on a white background" text suffixes trimmed
A Disney princess tattoo
A tattoo of Alice in wonderland falling down the rabbit hole
A tattoo of a psychedelic acid trip
A rose tattoo
A crazy clown tattoo
Rainbow unicorn tattoo
Rainbow unicorn tattoo
A female Night elf WoW tattoo
A male Night elf WoW tattoo
Base model
stabilityai/stable-diffusion-xl-base-1.0