import gradio as gr
from gradio_image_slider import ImageSlider
import numpy as np
import random
import torch
import spaces
from PIL import Image
from diffusers import FlowMatchEulerDiscreteScheduler
from optimization import optimize_pipeline_
from qwenimage.pipeline_qwenimage_edit_plus import QwenImageEditPlusPipeline
from qwenimage.transformer_qwenimage import QwenImageTransformer2DModel
from qwenimage.qwen_fa3_processor import QwenDoubleStreamAttnProcessorFA3
import math
from huggingface_hub import hf_hub_download
from safetensors.torch import load_file
from PIL import Image
import os
# --- Model Loading ---
dtype = torch.bfloat16
device = "cuda" if torch.cuda.is_available() else "cpu"
pipe = QwenImageEditPlusPipeline.from_pretrained("Qwen/Qwen-Image-Edit-2509",
transformer= QwenImageTransformer2DModel.from_pretrained("linoyts/Qwen-Image-Edit-Rapid-AIO",
subfolder='transformer',
torch_dtype=dtype,
device_map='cuda'),torch_dtype=dtype).to(device)
pipe.load_lora_weights("vafipas663/Qwen-Edit-2509-Upscale-LoRA",
weight_name="qwen-edit-enhance_64-v3_000001500.safetensors",
adapter_name="upscale")
pipe.set_adapters(["upscale"], adapter_weights=[1.])
pipe.fuse_lora(adapter_names=["upscale"], lora_scale=1.0)
pipe.unload_lora_weights()
pipe.transformer.__class__ = QwenImageTransformer2DModel
pipe.transformer.set_attn_processor(QwenDoubleStreamAttnProcessorFA3())
optimize_pipeline_(pipe, image=[Image.new("RGB", (1024, 1024)), Image.new("RGB", (1024, 1024))], prompt="prompt")
MAX_SEED = np.iinfo(np.int32).max
@spaces.GPU
def upscale_image(
image,
seed,
randomize_seed,
true_guidance_scale,
num_inference_steps,
height,
width,
progress=gr.Progress(track_tqdm=True)
):
prompt = "Upscale and enhance this image with high quality details"
if randomize_seed:
seed = random.randint(0, MAX_SEED)
generator = torch.Generator(device=device).manual_seed(seed)
pil_images = []
if image is not None:
if isinstance(image, Image.Image):
pil_images.append(image.convert("RGB"))
elif hasattr(image, "name"):
pil_images.append(Image.open(image.name).convert("RGB"))
if len(pil_images) == 0:
raise gr.Error("Please upload an image first.")
result = pipe(
image=pil_images,
prompt=prompt,
height=height if height != 0 else None,
width=width if width != 0 else None,
num_inference_steps=num_inference_steps,
generator=generator,
true_cfg_scale=true_guidance_scale,
num_images_per_prompt=1,
).images[0]
return (image, result), seed
# --- UI ---
css = '''
#col-container {
max-width: 900px;
margin: 0 auto;
padding: 2rem;
font-family: -apple-system, BlinkMacSystemFont, "Segoe UI", Roboto, Helvetica, Arial, sans-serif;
}
.gradio-container {
background: linear-gradient(to bottom, #f5f5f7, #ffffff);
}
#title {
text-align: center;
font-size: 2.5rem;
font-weight: 600;
color: #1d1d1f;
margin-bottom: 0.5rem;
}
#description {
text-align: center;
font-size: 1.1rem;
color: #6e6e73;
margin-bottom: 2rem;
}
.image-container {
border-radius: 18px;
overflow: hidden;
box-shadow: 0 4px 6px rgba(0, 0, 0, 0.07);
}
#convert-btn {
background: linear-gradient(180deg, #0071e3 0%, #0077ed 100%);
border: none;
border-radius: 12px;
color: white;
font-size: 1.1rem;
font-weight: 500;
padding: 0.75rem 2rem;
transition: all 0.3s ease;
}
#convert-btn:hover {
transform: translateY(-2px);
box-shadow: 0 8px 16px rgba(0, 113, 227, 0.3);
}
'''
def update_dimensions_on_upload(image):
if image is None:
return 1024, 1024
original_width, original_height = image.size
if original_width > original_height:
new_width = 1024
aspect_ratio = original_height / original_width
new_height = int(new_width * aspect_ratio)
else:
new_height = 1024
aspect_ratio = original_width / original_height
new_width = int(new_height * aspect_ratio)
# Ensure dimensions are multiples of 8
new_width = (new_width // 8) * 8
new_height = (new_height // 8) * 8
return new_width, new_height
with gr.Blocks(theme=gr.themes.Soft(), css=css) as demo:
with gr.Column(elem_id="col-container"):
gr.Markdown("# 🔍 Image Upscaler", elem_id="title")
gr.Markdown(
"""
Upscale and enhance your images with AI-powered quality improvement ✨
""",
elem_id="description"
)
with gr.Column():
image = gr.Image(
label="Upload Image",
type="pil",
elem_classes="image-container"
)
with gr.Accordion("⚙️ Advanced Settings", open=False):
seed = gr.Slider(label="Seed", minimum=0, maximum=MAX_SEED, step=1, value=0)
randomize_seed = gr.Checkbox(label="Randomize Seed", value=True)
true_guidance_scale = gr.Slider(label="Guidance Scale", minimum=1.0, maximum=10.0, step=0.1, value=1.0)
num_inference_steps = gr.Slider(label="Inference Steps", minimum=1, maximum=40, step=1, value=4)
height = gr.Slider(label="Height", minimum=256, maximum=2048, step=8, value=1024, visible=False)
width = gr.Slider(label="Width", minimum=256, maximum=2048, step=8, value=1024, visible=False)
upscale_btn = gr.Button("Upscale Image", variant="primary", elem_id="convert-btn", size="lg")
result = ImageSlider(
label="Before / After",
interactive=False,
elem_classes="image-container"
)
inputs = [
image, seed, randomize_seed, true_guidance_scale,
num_inference_steps, height, width
]
outputs = [result, seed]
# Upscale button click
upscale_btn.click(
fn=upscale_image,
inputs=inputs,
outputs=outputs
)
# Image upload triggers dimension update
image.upload(
fn=update_dimensions_on_upload,
inputs=[image],
outputs=[width, height]
)
demo.launch()