app.py

import os
import gradio as gr
import json
import logging
import torch
from PIL import Image
import spaces
from diffusers import DiffusionPipeline, FlowMatchEulerDiscreteScheduler
from huggingface_hub import hf_hub_download, HfFileSystem, ModelCard, snapshot_download
import copy
import random
import time
import re
import math
import numpy as np

# Load LoRAs from JSON
loras = [
    {
        "repo": "flymy-ai/qwen-image-realism-lora",
        "image": "https://huggingface.co/flymy-ai/qwen-image-realism-lora/resolve/main/assets/flymy_realism.png",
        "trigger_word": "Super Realism portrait of",
        "trigger_position": "prepend",
        "title": "Super Realism"
    },
    {
        "repo": "threecrowco/VolkClipartQwen",
        "image": "https://huggingface.co/threecrowco/VolkClipartQwen/resolve/main/images/_app_ai-toolkit_output_VolkDrawings_Qwen_v1_samples_1754805220500__000003000_3.jpg",
        "trigger_word": "volk clipart, black and white, ",
        "trigger_position": "prepend",
        "title": "Volk Clipart"
    },
    {
        "repo": "janekm/analog_film",
        "image": "https://huggingface.co/spaces/multimodalart/Qwen-Image-LoRA-Explorer/resolve/main/cat.webp",
        "trigger_word": "fifthel",
        "trigger_position": "prepend",
        "weights": "converted_complete.safetensors",
        "title": "Analog Film"
    },
    {
        "repo": "itspoidaman/qwenglitch",
        "image": "https://huggingface.co/itspoidaman/qwenglitch/resolve/main/images/GydaJ5LbEAAWKJU.jpeg",
        "trigger_word": "qwenglitch",
        "title": "Glitch"
    },
    {
        "repo": "alfredplpl/qwen-image-modern-anime-lora",
        "image": "https://huggingface.co/alfredplpl/qwen-image-modern-anime-lora/resolve/main/sample1.jpg",
        "trigger_word": "Japanese modern anime style, ",
        "trigger_position": "prepend",
        "title": "Modern Anime"
    },
    {
        "repo": "lichorosario/qwen-image-dottrmstr",
        "image": "https://huggingface.co/lichorosario/qwen-image-dottrmstr/resolve/main/images/Day_of_the_Tentacle_Remastered_(PC)_08.jpg",
        "trigger_word": "DOTTRMSTR",
        "trigger_position": "prepend",
        "title": "Day of the Tentacle Style"
    }
]

# Initialize the base model
dtype = torch.bfloat16
device = "cuda" if torch.cuda.is_available() else "cpu"
base_model = "Qwen/Qwen-Image"

# Scheduler configuration from the Qwen-Image-Lightning repository
scheduler_config = {
    "base_image_seq_len": 256,
    "base_shift": math.log(3),
    "invert_sigmas": False,
    "max_image_seq_len": 8192,
    "max_shift": math.log(3),
    "num_train_timesteps": 1000,
    "shift": 1.0,
    "shift_terminal": None,
    "stochastic_sampling": False,
    "time_shift_type": "exponential",
    "use_beta_sigmas": False,
    "use_dynamic_shifting": True,
    "use_exponential_sigmas": False,
    "use_karras_sigmas": False,
}

scheduler = FlowMatchEulerDiscreteScheduler.from_config(scheduler_config)
pipe = DiffusionPipeline.from_pretrained(
    base_model, scheduler=scheduler, torch_dtype=dtype
).to(device)

# Lightning LoRA info (no global state)
LIGHTNING_LORA_REPO = "lightx2v/Qwen-Image-Lightning"
LIGHTNING_LORA_WEIGHT = "Qwen-Image-Lightning-8steps-V1.0.safetensors"

MAX_SEED = np.iinfo(np.int32).max

class calculateDuration:
    def __init__(self, activity_name=""):
        self.activity_name = activity_name

    def __enter__(self):
        self.start_time = time.time()
        return self
    
    def __exit__(self, exc_type, exc_value, traceback):
        self.end_time = time.time()
        self.elapsed_time = self.end_time - self.start_time
        if self.activity_name:
            print(f"Elapsed time for {self.activity_name}: {self.elapsed_time:.6f} seconds")
        else:
            print(f"Elapsed time: {self.elapsed_time:.6f} seconds")

def get_image_size(aspect_ratio):
    """Converts aspect ratio string to width, height tuple."""
    if aspect_ratio == "1:1":
        return 1024, 1024
    elif aspect_ratio == "16:9":
        return 1152, 640
    elif aspect_ratio == "9:16":
        return 640, 1152
    elif aspect_ratio == "4:3":
        return 1024, 768
    elif aspect_ratio == "3:4":
        return 768, 1024
    elif aspect_ratio == "3:2":
        return 1024, 688
    elif aspect_ratio == "2:3":
        return 688, 1024
    else:
        return 1024, 1024

def update_selection(evt: gr.SelectData, aspect_ratio):
    selected_lora = loras[evt.index]
    new_placeholder = f"Type a prompt for {selected_lora['title']}"
    lora_repo = selected_lora["repo"]
    updated_text = f"### Selected: [{lora_repo}](https://huggingface.co/{lora_repo}) ✨"
    
    # Update aspect ratio if specified in LoRA config
    if "aspect" in selected_lora:
        if selected_lora["aspect"] == "portrait":
            aspect_ratio = "9:16"
        elif selected_lora["aspect"] == "landscape":
            aspect_ratio = "16:9"
        else:
            aspect_ratio = "1:1"
    
    return (
        gr.update(placeholder=new_placeholder),
        updated_text,
        evt.index,
        aspect_ratio,
    )

def handle_speed_mode(speed_mode):
    """Update UI based on speed/quality toggle."""
    if speed_mode == "Speed (8 steps)":
        return gr.update(value="Speed mode selected - 8 steps with Lightning LoRA"), 8, 1.0
    else: 
        return gr.update(value="Quality mode selected - 45 steps for best quality"), 45, 3.5

@spaces.GPU(duration=70)
def generate_image(prompt_mash, steps, seed, cfg_scale, width, height, lora_scale, negative_prompt=""):
    pipe.to("cuda")
    generator = torch.Generator(device="cuda").manual_seed(seed)
    
    with calculateDuration("Generating image"):
        # Generate image
        image = pipe(
            prompt=prompt_mash,
            negative_prompt=negative_prompt,
            num_inference_steps=steps,
            true_cfg_scale=cfg_scale,  # Use true_cfg_scale for Qwen-Image
            width=width,
            height=height,
            generator=generator,
        ).images[0]
        
    return image

@spaces.GPU(duration=70)
def run_lora(prompt, cfg_scale, steps, selected_index, randomize_seed, seed, aspect_ratio, lora_scale, speed_mode, progress=gr.Progress(track_tqdm=True)):
    if selected_index is None:
        raise gr.Error("You must select a LoRA before proceeding.")
    
    selected_lora = loras[selected_index]
    lora_path = selected_lora["repo"]
    trigger_word = selected_lora["trigger_word"]
    
    # Prepare prompt with trigger word
    if trigger_word:
        if "trigger_position" in selected_lora:
            if selected_lora["trigger_position"] == "prepend":
                prompt_mash = f"{trigger_word} {prompt}"
            else:
                prompt_mash = f"{prompt} {trigger_word}"
        else:
            prompt_mash = f"{trigger_word} {prompt}"
    else:
        prompt_mash = prompt

    # Always unload any existing LoRAs first to avoid conflicts
    with calculateDuration("Unloading existing LoRAs"):
        pipe.unload_lora_weights()

    # Load LoRAs based on speed mode
    if speed_mode == "Speed (8 steps)":
        with calculateDuration("Loading Lightning LoRA and style LoRA"):
            # Load Lightning LoRA first
            pipe.load_lora_weights(
                LIGHTNING_LORA_REPO, 
                weight_name=LIGHTNING_LORA_WEIGHT,
                adapter_name="lightning"
            )
            
            # Load the selected style LoRA
            weight_name = selected_lora.get("weights", None)
            pipe.load_lora_weights(
                lora_path, 
                weight_name=weight_name, 
                low_cpu_mem_usage=True,
                adapter_name="style"
            )
            
            # Set both adapters active with their weights
            pipe.set_adapters(["lightning", "style"], adapter_weights=[1.0, lora_scale])
    else:
        # Quality mode - only load the style LoRA
        with calculateDuration(f"Loading LoRA weights for {selected_lora['title']}"):
            weight_name = selected_lora.get("weights", None)
            pipe.load_lora_weights(
                lora_path, 
                weight_name=weight_name, 
                low_cpu_mem_usage=True
            )
                
    # Set random seed for reproducibility
    with calculateDuration("Randomizing seed"):
        if randomize_seed:
            seed = random.randint(0, MAX_SEED)
    
    # Get image dimensions from aspect ratio
    width, height = get_image_size(aspect_ratio)
    
    # Generate the image
    final_image = generate_image(prompt_mash, steps, seed, cfg_scale, width, height, lora_scale)
    
    return final_image, seed

def get_huggingface_safetensors(link):
    split_link = link.split("/")
    if len(split_link) != 2:
        raise Exception("Invalid Hugging Face repository link format.")

    print(f"Repository attempted: {split_link}")
    
    # Load model card
    model_card = ModelCard.load(link)
    base_model = model_card.data.get("base_model")
    print(f"Base model: {base_model}")

    # Validate model type (for Qwen-Image)
    acceptable_models = {"Qwen/Qwen-Image"}
    
    models_to_check = base_model if isinstance(base_model, list) else [base_model]
    
    if not any(model in acceptable_models for model in models_to_check):
        raise Exception("Not a Qwen-Image LoRA!")
        
    # Extract image and trigger word
    image_path = model_card.data.get("widget", [{}])[0].get("output", {}).get("url", None)
    trigger_word = model_card.data.get("instance_prompt", "")
    image_url = f"https://huggingface.co/{link}/resolve/main/{image_path}" if image_path else None

    # Initialize Hugging Face file system
    fs = HfFileSystem()
    try:
        list_of_files = fs.ls(link, detail=False)
        
        # Find safetensors file
        safetensors_name = None
        for file in list_of_files:
            filename = file.split("/")[-1]
            if filename.endswith(".safetensors"):
                safetensors_name = filename
                break

        if not safetensors_name:
            raise Exception("No valid *.safetensors file found in the repository.")

    except Exception as e:
        print(e)
        raise Exception("You didn't include a valid Hugging Face repository with a *.safetensors LoRA")
    
    return split_link[1], link, safetensors_name, trigger_word, image_url

def check_custom_model(link):
    print(f"Checking a custom model on: {link}")
    
    if link.endswith('.safetensors'):
        if 'huggingface.co' in link:
            parts = link.split('/')
            try:
                hf_index = parts.index('huggingface.co')
                username = parts[hf_index + 1]
                repo_name = parts[hf_index + 2]
                repo = f"{username}/{repo_name}"
                
                safetensors_name = parts[-1]
                
                try:
                    model_card = ModelCard.load(repo)
                    trigger_word = model_card.data.get("instance_prompt", "")
                    image_path = model_card.data.get("widget", [{}])[0].get("output", {}).get("url", None)
                    image_url = f"https://huggingface.co/{repo}/resolve/main/{image_path}" if image_path else None
                except:
                    trigger_word = ""
                    image_url = None
                
                return repo_name, repo, safetensors_name, trigger_word, image_url
            except:
                raise Exception("Invalid safetensors URL format")
    
    if link.startswith("https://"):
        if link.startswith("https://huggingface.co") or link.startswith("https://www.huggingface.co"):
            link_split = link.split("huggingface.co/")
            return get_huggingface_safetensors(link_split[1])
    else: 
        return get_huggingface_safetensors(link)

def add_custom_lora(custom_lora):
    global loras
    if custom_lora:
        try:
            title, repo, path, trigger_word, image = check_custom_model(custom_lora)
            print(f"Loaded custom LoRA: {repo}")
            card = f'''
            <div class="custom_lora_card">
              <span>Loaded custom LoRA:</span>
              <div class="card_internal">
                <img src="{image}" />
                <div>
                    <h3>{title}</h3>
                    <small>{"Using: <code><b>"+trigger_word+"</code></b> as the trigger word" if trigger_word else "No trigger word found. If there's a trigger word, include it in your prompt"}<br></small>
                </div>
              </div>
            </div>
            '''
            existing_item_index = next((index for (index, item) in enumerate(loras) if item['repo'] == repo), None)
            if existing_item_index is None:
                new_item = {
                    "image": image,
                    "title": title,
                    "repo": repo,
                    "weights": path,
                    "trigger_word": trigger_word
                }
                print(new_item)
                loras.append(new_item)
                existing_item_index = len(loras) - 1  # Get the actual index after adding
        
            return gr.update(visible=True, value=card), gr.update(visible=True), gr.Gallery(selected_index=None), f"Custom: {path}", existing_item_index, trigger_word
        except Exception as e:
            gr.Warning(f"Invalid LoRA: either you entered an invalid link, or a non-Qwen-Image LoRA, this was the issue: {e}")
            return gr.update(visible=True, value=f"Invalid LoRA: either you entered an invalid link, a non-Qwen-Image LoRA"), gr.update(visible=True), gr.update(), "", None, ""
    else:
        return gr.update(visible=False), gr.update(visible=False), gr.update(), "", None, ""

def remove_custom_lora():
    return gr.update(visible=False), gr.update(visible=False), gr.update(), "", None, ""

run_lora.zerogpu = True

css = '''
#gen_btn{height: 100%}
#gen_column{align-self: stretch}
#title{text-align: center}
#title h1{font-size: 3em; display:inline-flex; align-items:center}
#title img{width: 100px; margin-right: 0.5em}
#gallery .grid-wrap{height: 10vh}
#lora_list{background: var(--block-background-fill);padding: 0 1em .3em; font-size: 90%}
.card_internal{display: flex;height: 100px;margin-top: .5em}
.card_internal img{margin-right: 1em}
.styler{--form-gap-width: 0px !important}
#speed_status{padding: .5em; border-radius: 5px; margin: 1em 0}
'''

with gr.Blocks(theme=gr.themes.Soft(), css=css, delete_cache=(60, 60)) as app:
    title = gr.HTML(
        """<img src="https://qianwen-res.oss-cn-beijing.aliyuncs.com/Qwen-Image/qwen_image_logo.png" alt="Qwen-Image" style="width: 280px; margin: 0 auto">        
        <h3 style="margin-top: -10px">LoRA Explorer</h3>""",
        elem_id="title",
    )
    
    selected_index = gr.State(None)
    
    with gr.Row():
        with gr.Column(scale=3):
            prompt = gr.Textbox(label="Prompt", lines=1, placeholder="Type a prompt after selecting a LoRA")
        with gr.Column(scale=1, elem_id="gen_column"):
            generate_button = gr.Button("Generate", variant="primary", elem_id="gen_btn")
    
    with gr.Row():
        with gr.Column():
            selected_info = gr.Markdown("")
            gallery = gr.Gallery(
                [(item["image"], item["title"]) for item in loras],
                label="LoRA Gallery",
                allow_preview=False,
                columns=3,
                elem_id="gallery",
                show_share_button=False
            )
            with gr.Group():
                custom_lora = gr.Textbox(label="Custom LoRA", info="LoRA Hugging Face path", placeholder="username/qwen-image-custom-lora")
                gr.Markdown("[Check Qwen-Image LoRAs](https://huggingface.co/models?other=base_model:adapter:Qwen/Qwen-Image)", elem_id="lora_list")
            custom_lora_info = gr.HTML(visible=False)
            custom_lora_button = gr.Button("Remove custom LoRA", visible=False)
        
        with gr.Column():
            result = gr.Image(label="Generated Image")
            
            with gr.Row():
                speed_mode = gr.Radio(
                    label="Generation Mode",
                    choices=["Speed (8 steps)", "Quality (45 steps)"],
                    value="Quality (45 steps)",
                    info="Speed mode uses Lightning LoRA for faster generation"
                )
            
            speed_status = gr.Markdown("Quality mode active", elem_id="speed_status")

    with gr.Row():
        with gr.Accordion("Advanced Settings", open=False):
            with gr.Column():
                with gr.Row():
                    aspect_ratio = gr.Radio(
                        label="Aspect Ratio",
                        choices=["1:1", "16:9", "9:16", "4:3", "3:4", "3:2", "2:3"],
                        value="1:1"
                    )
                    
                with gr.Row():
                    cfg_scale = gr.Slider(
                        label="Guidance Scale (True CFG)", 
                        minimum=1.0, 
                        maximum=5.0, 
                        step=0.1, 
                        value=3.5,
                        info="Lower for speed mode, higher for quality"
                    )
                    steps = gr.Slider(
                        label="Steps", 
                        minimum=4, 
                        maximum=50, 
                        step=1, 
                        value=45,
                        info="Automatically set by speed mode"
                    )
                
                with gr.Row():
                    randomize_seed = gr.Checkbox(True, label="Randomize seed")
                    seed = gr.Slider(label="Seed", minimum=0, maximum=MAX_SEED, step=1, value=0, randomize=True)
                    lora_scale = gr.Slider(label="LoRA Scale", minimum=0, maximum=2, step=0.01, value=1.0)

    # Event handlers
    gallery.select(
        update_selection,
        inputs=[aspect_ratio],
        outputs=[prompt, selected_info, selected_index, aspect_ratio]
    )
    
    speed_mode.change(
        handle_speed_mode,
        inputs=[speed_mode],
        outputs=[speed_status, steps, cfg_scale]
    )
    
    custom_lora.input(
        add_custom_lora,
        inputs=[custom_lora],
        outputs=[custom_lora_info, custom_lora_button, gallery, selected_info, selected_index, prompt]
    )
    
    custom_lora_button.click(
        remove_custom_lora,
        outputs=[custom_lora_info, custom_lora_button, gallery, selected_info, selected_index, custom_lora]
    )
    
    gr.on(
        triggers=[generate_button.click, prompt.submit],
        fn=run_lora,
        inputs=[prompt, cfg_scale, steps, selected_index, randomize_seed, seed, aspect_ratio, lora_scale, speed_mode],
        outputs=[result, seed]
    )

app.queue()
app.launch()