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("JCTN/JCTN_LORAxl")
prompt = " boogers, free tissue inside"
image = pipe(prompt).images[0]Super Cereal - SDXL LoRA
boogers, free tissue inside
Multiplier of 0.9 - 1.1 works well on SDXL base. Simple prompts tend to work well. No trigger word needed.
Special thanks to Huggingface for the GPU grant.
Image examples for the model:
star wars wookie bits, free lightsaber inside
kitty litter crunch
t bone steak
black plague, free death inside
barbie and ken
boiled eggs
raw bacon
herpes
pickles
- Downloads last month
- 5,146
Model tree for JCTN/JCTN_LORAxl
Base model
stabilityai/stable-diffusion-xl-base-1.0