This is a simple code that uses stable diffusion XL for text guided image generation. Check the official SDXL huggingface page for more details.
A demo image:
Generated by:
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
from diffusers import DiffusionPipeline
import torch
import argparse
import os
# load both base & refiner
base = DiffusionPipeline.from_pretrained(
"stabilityai/stable-diffusion-xl-base-1.0", torch_dtype=torch.float16, variant="fp16", use_safetensors=True
)
base.to("cuda")
refiner = DiffusionPipeline.from_pretrained(
"stabilityai/stable-diffusion-xl-refiner-1.0",
text_encoder_2=base.text_encoder_2,
vae=base.vae,
torch_dtype=torch.float16,
use_safetensors=True,
variant="fp16",
)
refiner.to("cuda")
def SDXL_img(prompt, n_steps=40, high_noise_frac=0.8):
# run both experts
image = base(
prompt=prompt,
num_inference_steps=n_steps,
denoising_end=high_noise_frac,
output_type="latent",
).images
image = refiner(
prompt=prompt,
num_inference_steps=n_steps,
denoising_start=high_noise_frac,
image=image,
).images[0]
return image
if __name__ == "__main__":
save_folder = './SDXL_img'
file_name = 'golden_retriver.png'
save_dir = os.path.join(save_folder, file_name)
prompt = "A cute golden retriver puppy with background of mapple leaves."
image = SDXL_img(prompt)
image.save(save_dir)