app.py

import gradio as gr
import json
import logging
import torch
from PIL import Image
import spaces
from diffusers import DiffusionPipeline
import copy

# Load LoRAs from JSON file
with open('loras.json', 'r') as f:
    loras = json.load(f)

# Initialize the base model
base_model = "black-forest-labs/FLUX.1-dev"
pipe = DiffusionPipeline.from_pretrained(base_model, torch_dtype=torch.bfloat16)
pipe.to("cuda")

MAX_SEED = 2**32-1

def update_selection(evt: gr.SelectData):
    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}) ✨"
    return (
        gr.update(placeholder=new_placeholder),
        updated_text,
        evt.index
    )

@spaces.GPU(duration=90)
def run_lora(prompt, cfg_scale, steps, selected_index, randomize_seed, seed, width, height, lora_scale, 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"]

    # Load LoRA weights
    if "weights" in selected_lora:
        pipe.load_lora_weights(lora_path, weight_name=selected_lora["weights"])
    else:
        pipe.load_lora_weights(lora_path)
        
    # Set random seed for reproducibility
    if randomize_seed:
        seed = random.randint(0, MAX_SEED)
    generator = torch.Generator(device="cuda").manual_seed(seed)

    # Generate image
    image = pipe(
        prompt=f"{prompt} {trigger_word}",
        num_inference_steps=steps,
        guidance_scale=cfg_scale,
        width=width,
        height=height,
        generator=generator,
        joint_attention_kwargs={"scale": lora_scale},
    ).images[0]

    yield image
    
    pipe.unload_lora_weights()
    

'''
#gen_column{height: 100%}
#gen_btn{height: 100%}
'''
with gr.Blocks(theme=gr.themes.Soft()) as app:
    gr.Markdown("# FLUX.1 LoRA the Explorer")
    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(scale=3):
            selected_info = gr.Markdown("")
            gallery = gr.Gallery(
                [(item["image"], item["title"]) for item in loras],
                label="LoRA Gallery",
                allow_preview=False,
                columns=3
            )
            
        with gr.Column(scale=4):
            result = gr.Image(label="Generated Image")

    with gr.Row():
        with gr.Accordion("Advanced Settings", open=False):
            with gr.Column():
                with gr.Row():
                    cfg_scale = gr.Slider(label="CFG Scale", minimum=1, maximum=20, step=0.5, value=3.5)
                    steps = gr.Slider(label="Steps", minimum=1, maximum=100, step=1, value=30)
                
                with gr.Row():
                    width = gr.Slider(label="Width", minimum=256, maximum=1536, step=64, value=1024)
                    height = gr.Slider(label="Height", minimum=256, maximum=1536, step=64, value=1024)
                
                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=1, step=0.01, value=0.85)

    gallery.select(update_selection, outputs=[prompt, selected_info, selected_index])

    gr.on(
        triggers=[generate_button.click, prompt.submit],
        fn=run_lora,
        inputs=[prompt, cfg_scale, steps, selected_index, randomize_seed, seed, width, height, lora_scale],
        outputs=[result]
    )

app.queue()
app.launch()