How to use pskl/icondiffusion with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("pskl/icondiffusion", dtype=torch.bfloat16, device_map="cuda") prompt = "asim style, a tree, centered, monochrome, line-art, logo" image = pipe(prompt).images[0]
SD Offset noise 1.5 finetuned on 512x512 svgs.
