Downloads · 30 days
16
0% of all-time downloads
UCSC-VLAA/HQ-Edit-ckpt
HQ-Edit-ckpt is a text-to-image model from UCSC-VLAA. Use it when you need an image from a text prompt. It is set up for diffusers.
Make sure to install the libraries first:
Downloads · 30 days
16
0% of all-time downloads
All-time downloads
8.8K
Public
Parameters
860M
15.4 GB on disk
Likes
13
Public
Click a slice to open those files.
.safetensors11.9 GB · 63%
From the Hugging Face model README
Make sure to install the libraries first:
pip install accelerate transformers
pip install git+https://github.com/huggingface/diffusers
import torch
import torch
from diffusers import StableDiffusionInstructPix2PixPipeline, EulerAncestralDiscreteScheduler
from diffusers.utils import load_image
image_guidance_scale = 1.5
guidance_scale = 7.0
model_id = "MudeHui/HQ-Edit"
pipe = StableDiffusionInstructPix2PixPipeline.from_pretrained(model_id, torch_dtype=torch.float16, safety_checker=None)
pipe.to("cuda")
pipe.scheduler = EulerAncestralDiscreteScheduler.from_config(pipe.scheduler.config)
resolution = 512
image = load_image(
"https://hf.co/datasets/diffusers/diffusers-images-docs/resolve/main/mountain.png"
).resize((resolution, resolution))
edit_instruction = "Turn sky into a cloudy one"
edited_image = pipe(
prompt=edit_instruction,
image=image,
height=resolution,
width=resolution,
guidance_scale=image_guidance_scale,
image_guidance_scale=image_guidance_scale,
num_inference_steps=30,
).images[0]
edited_image.save("edited_image.png")