app.py

import gradio as gr
import numpy as np
import random
import torch
import spaces
from PIL import Image
from huggingface_hub import hf_hub_download
from safetensors.torch import load_file
from tqdm import tqdm
import gc

from qwenimage.pipeline_qwen_image_edit import QwenImageEditPipeline
from qwenimage.transformer_qwenimage import QwenImageTransformer2DModel
from qwenimage.qwen_fa3_processor import QwenDoubleStreamAttnProcessorFA3


LORA_CONFIG = {
    "None": {
        "repo_id": None,
        "filename": None,
        "type": "edit",
        "method": "none",
        "prompt_template": "{prompt}",
        "description": "Use the base Qwen-Image-Edit model without any LoRA.",
    },
    "InStyle (Style Transfer)": {
        "repo_id": "peteromallet/Qwen-Image-Edit-InStyle",
        "filename": "InStyle-0.5.safetensors",
        "type": "style",
        "method": "manual_fuse",
        "prompt_template": "Make an image in this style of {prompt}",
        "description": "Transfers the style from a reference image to a new image described by the prompt.",
    },
    "InScene (In-Scene Editing)": {
        "repo_id": "flymy-ai/qwen-image-edit-inscene-lora",
        "filename": "flymy_qwen_image_edit_inscene_lora.safetensors",
        "type": "edit",
        "method": "standard",
        "prompt_template": "{prompt}",
        "description": "Improves in-scene editing, object positioning, and camera perspective changes.",
    },
    "Face Segmentation": {
        "repo_id": "TsienDragon/qwen-image-edit-lora-face-segmentation",
        "filename": "pytorch_lora_weights.safetensors",
        "type": "edit",
        "method": "standard",
        "prompt_template": "change the face to face segmentation mask",
        "description": "Transforms a facial image into a precise segmentation mask.",
    },
    "Object Remover": {
        "repo_id": "valiantcat/Qwen-Image-Edit-Remover-General-LoRA",
        "filename": "qwen-edit-remover.safetensors",
        "type": "edit",
        "method": "standard",
        "prompt_template": "Remove {prompt}",
        "description": "Removes objects from an image while maintaining background consistency.",
    },
}

print("Initializing model...")
dtype = torch.bfloat16
device = "cuda" if torch.cuda.is_available() else "cpu"

pipe = QwenImageEditPipeline.from_pretrained(
    "Qwen/Qwen-Image-Edit", 
    torch_dtype=dtype
).to(device)

pipe.transformer.__class__ = QwenImageTransformer2DModel
pipe.transformer.set_attn_processor(QwenDoubleStreamAttnProcessorFA3())

original_transformer_state_dict = pipe.transformer.state_dict()
print("Base model loaded and ready.")

def fuse_lora_manual(transformer, lora_state_dict, alpha=1.0):
    key_mapping = {}
    for key in lora_state_dict.keys():
        base_key = key.replace('diffusion_model.', '').rsplit('.lora_', 1)[0]
        if base_key not in key_mapping:
            key_mapping[base_key] = {}
        if 'lora_A' in key:
            key_mapping[base_key]['down'] = lora_state_dict[key]
        elif 'lora_B' in key:
            key_mapping[base_key]['up'] = lora_state_dict[key]

    for name, module in tqdm(transformer.named_modules(), desc="Fusing layers"):
        if name in key_mapping and isinstance(module, torch.nn.Linear):
            lora_weights = key_mapping[name]
            if 'down' in lora_weights and 'up' in lora_weights:
                device = module.weight.device
                dtype = module.weight.dtype
                lora_down = lora_weights['down'].to(device, dtype=dtype)
                lora_up = lora_weights['up'].to(device, dtype=dtype)
                merged_delta = lora_up @ lora_down
                module.weight.data += alpha * merged_delta
    return transformer

def load_and_fuse_lora(lora_name):
    """Carrega uma LoRA, funde-a ao modelo e retorna o pipeline modificado."""
    config = LORA_CONFIG[lora_name]
    
    print("Resetting transformer to original state...")
    pipe.transformer.load_state_dict(original_transformer_state_dict)
    
    if config["method"] == "none":
        print("No LoRA selected. Using base model.")
        return

    print(f"Loading LoRA: {lora_name}")
    lora_path = hf_hub_download(repo_id=config["repo_id"], filename=config["filename"])

    if config["method"] == "standard":
        print("Using standard loading method...")
        pipe.load_lora_weights(lora_path)
        print("Fusing LoRA into the model...")
        pipe.fuse_lora()
    elif config["method"] == "manual_fuse":
        print("Using manual fusion method...")
        lora_state_dict = load_file(lora_path)
        pipe.transformer = fuse_lora_manual(pipe.transformer, lora_state_dict)
    
    gc.collect()
    torch.cuda.empty_cache()
    print(f"LoRA '{lora_name}' is now active.")

@spaces.GPU(duration=60)
def infer(
    lora_name,
    input_image,
    style_image,
    prompt,
    seed,
    randomize_seed,
    true_guidance_scale,
    num_inference_steps,
    progress=gr.Progress(track_tqdm=True),
):
    if not lora_name:
        raise gr.Error("Please select a LoRA model.")
    
    config = LORA_CONFIG[lora_name]
    
    if config["type"] == "style":
        if style_image is None:
            raise gr.Error("Style Transfer LoRA requires a Style Reference Image.")
        image_for_pipeline = style_image
    else: # 'edit'
        if input_image is None:
            raise gr.Error("This LoRA requires an Input Image.")
        image_for_pipeline = input_image

    if not prompt and config["prompt_template"] != "change the face to face segmentation mask":
        raise gr.Error("A text prompt is required for this LoRA.")
    
    load_and_fuse_lora(lora_name)
    
    final_prompt = config["prompt_template"].format(prompt=prompt)
    
    if randomize_seed:
        seed = random.randint(0, np.iinfo(np.int32).max)
    generator = torch.Generator(device=device).manual_seed(int(seed))
    
    print("--- Running Inference ---")
    print(f"LoRA: {lora_name}")
    print(f"Prompt: {final_prompt}")
    print(f"Seed: {seed}, Steps: {num_inference_steps}, CFG: {true_guidance_scale}")
    
    with torch.inference_mode():
        result_image = pipe(
            image=image_for_pipeline,
            prompt=final_prompt,
            negative_prompt=" ",
            num_inference_steps=int(num_inference_steps),
            generator=generator,
            true_cfg_scale=true_guidance_scale,
        ).images[0]
        
    pipe.unfuse_lora()
    gc.collect()
    torch.cuda.empty_cache()
    
    return result_image, seed

def on_lora_change(lora_name):
    config = LORA_CONFIG[lora_name]
    is_style_lora = config["type"] == "style"
    return {
        lora_description: gr.Markdown(visible=True, value=f"**Description:** {config['description']}"),
        input_image_box: gr.Image(visible=not is_style_lora),
        style_image_box: gr.Image(visible=is_style_lora),
        prompt_box: gr.Textbox(visible=(config["prompt_template"] != "change the face to face segmentation mask"))
    }

with gr.Blocks(css="#col-container { margin: 0 auto; max-width: 1024px; }") as demo:
    with gr.Column(elem_id="col-container"):
        gr.HTML('<img src="https://qianwen-res.oss-cn-beijing.aliyuncs.com/Qwen-Image/qwen_image_edit_logo.png" alt="Qwen-Image Logo" style="width: 400px; margin: 0 auto; display: block;">')
        gr.Markdown("<h2 style='text-align: center;'>Qwen-Image-Edit Multi-LoRA Playground</h2>")

        with gr.Row():
            with gr.Column(scale=1):
                lora_selector = gr.Dropdown(
                    label="Select LoRA Model",
                    choices=list(LORA_CONFIG.keys()),
                    value="InStyle (Style Transfer)"
                )
                lora_description = gr.Markdown(visible=False)
                
                input_image_box = gr.Image(label="Input Image", type="pil", visible=False)
                style_image_box = gr.Image(label="Style Reference Image", type="pil", visible=True)
                
                prompt_box = gr.Textbox(label="Prompt", placeholder="Describe the content or object to remove...")
                
                run_button = gr.Button("Generate!", variant="primary")

            with gr.Column(scale=1):
                result_image = gr.Image(label="Result", type="pil")
                used_seed = gr.Number(label="Used Seed", interactive=False)

        with gr.Accordion("Advanced Settings", open=False):
            seed_slider = gr.Slider(label="Seed", minimum=0, maximum=np.iinfo(np.int32).max, step=1, value=42)
            randomize_seed_checkbox = gr.Checkbox(label="Randomize seed", value=True)
            cfg_slider = gr.Slider(label="Guidance Scale (CFG)", minimum=1.0, maximum=10.0, step=0.1, value=4.0)
            steps_slider = gr.Slider(label="Inference Steps", minimum=10, maximum=50, step=1, value=25)

        lora_selector.change(
            fn=on_lora_change,
            inputs=lora_selector,
            outputs=[lora_description, input_image_box, style_image_box, prompt_box]
        )

        demo.load(
            fn=on_lora_change,
            inputs=lora_selector,
            outputs=[lora_description, input_image_box, style_image_box, prompt_box]
        )

        run_button.click(
            fn=infer,
            inputs=[
                lora_selector,
                input_image_box, style_image_box,
                prompt_box,
                seed_slider, randomize_seed_checkbox,
                cfg_slider, steps_slider
            ],
            outputs=[result_image, used_seed]
        )

if __name__ == "__main__":
    demo.launch()