tools / README.md
patrickvonplaten's picture
all
6c27fdd
|
raw
history blame
1.81 kB
metadata
language:
  - en
license: creativeml-openrail-m
tags:
  - stable-diffusion
  - stable-diffusion-diffusers
  - inpainting
  - art
  - artistic
  - diffusers
  - anime
  - absolute-realism

Absolute realism 1.6525 inpainting

lykon-absolute-realism/absolute-realism-1.6525-inpainting is a Stable Diffusion Inpainting model that has been fine-tuned on runwayml/stable-diffusion-inpainting.

Please consider supporting me:

Diffusers

For more general information on how to run inpainting models with 🧨 Diffusers, see the docs.

  1. Installation
pip install diffusers transformers accelerate
  1. Run
from diffusers import AutoPipelineForInpainting, DEISMultistepScheduler
import torch
from diffusers.utils import load_image

pipe = AutoPipelineForInpainting.from_pretrained('lykon-absolute-realism/absolute-realism-1.6525-inpainting', torch_dtype=torch.float16, variant="fp16")
pipe.scheduler = DEISMultistepScheduler.from_config(pipe.scheduler.config)
pipe = pipe.to("cuda")

img_url = "https://raw.githubusercontent.com/CompVis/latent-diffusion/main/data/inpainting_examples/overture-creations-5sI6fQgYIuo.png"
mask_url = "https://raw.githubusercontent.com/CompVis/latent-diffusion/main/data/inpainting_examples/overture-creations-5sI6fQgYIuo_mask.png"

image = load_image(img_url)
mask_image = load_image(mask_url)


prompt = "a majestic tiger sitting on a park bench"

generator = torch.manual_seed(33)
image = pipe(prompt, image=image, mask_image=mask_image, generator=generator, num_inference_steps=25).images[0]  
image.save("./image.png")