import torch
from diffusers import DiffusionPipeline
# switch to "mps" for apple devices
pipe = DiffusionPipeline.from_pretrained("fill-in-base-model", dtype=torch.bfloat16, device_map="cuda")
pipe.load_lora_weights("calm-and-collected/knives_and_time")
prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k"
image = pipe(prompt).images[0]Knives and time, a damaged LORA
Knives and time is a LORA model trained on public domain and CC0 images. The data was collected from wikimedia commons and annotated with a mixture of BLIP annotation and manual annotation.
Usage
The base trigger is katabd.
For paintings: painting, yellowed, abrasion, scratches, flaking, damage, torn, cut, pierced. For photography: photograph, monochrome, overlit, overexposed, torn, cut.
I recommend using it combined with an LCM sampler.
Training
The model was trained on the 'knives and time, a damaged dataset' for ~45 minutes resulting in six epochs.
- Downloads last month
- 11