app.py

import os
import sys
import random
import torch
from pathlib import Path
from PIL import Image
import gradio as gr
from huggingface_hub import hf_hub_download
import spaces
from typing import Union, Sequence, Mapping, Any

# Configuração inicial e diagnóstico CUDA
print("Python version:", sys.version)
print("Torch version:", torch.__version__)
print("CUDA disponível:", torch.cuda.is_available())
print("Quantidade de GPUs:", torch.cuda.device_count())
if torch.cuda.is_available():
    print("GPU atual:", torch.cuda.get_device_name(0))
else:
    print("CUDA não está disponível. Verificando por que:")
    try:
        torch.cuda.init()
    except Exception as e:
        print("Erro ao inicializar CUDA:", str(e))

# Adicionar o caminho da pasta ComfyUI ao sys.path
current_dir = os.path.dirname(os.path.abspath(__file__))
comfyui_path = os.path.join(current_dir, "ComfyUI")
sys.path.append(comfyui_path)

from nodes import NODE_CLASS_MAPPINGS
from comfy import model_management
import folder_paths

# Configuração de diretórios
BASE_DIR = os.path.dirname(os.path.realpath(__file__))
output_dir = os.path.join(BASE_DIR, "output")
os.makedirs(output_dir, exist_ok=True)
folder_paths.set_output_directory(output_dir)

# Helper function
def get_value_at_index(obj: Union[Sequence, Mapping], index: int) -> Any:
    try:
        return obj[index]
    except KeyError:
        return obj["result"][index]

# Baixar modelos necessários
def download_models():
    models = [
        ("black-forest-labs/FLUX.1-Redux-dev", "flux1-redux-dev.safetensors", "models/style_models"),
        ("comfyanonymous/flux_text_encoders", "t5xxl_fp16.safetensors", "models/text_encoders"),
        ("zer0int/CLIP-GmP-ViT-L-14", "ViT-L-14-TEXT-detail-improved-hiT-GmP-HF.safetensors", "models/text_encoders"),
        ("black-forest-labs/FLUX.1-dev", "ae.safetensors", "models/vae"),
        ("black-forest-labs/FLUX.1-dev", "flux1-dev.safetensors.safetensors", "models/diffusion_models"),
        ("google/siglip-so400m-patch14-384", "model.safetensors", "models/clip_vision"),
        ("nftnik/NFTNIK-FLUX.1-dev-LoRA", "NFTNIK_FLUX.1[dev]_LoRA.safetensors", "models/lora")
    ]
    
    for repo_id, filename, local_dir in models:
        hf_hub_download(repo_id=repo_id, filename=filename, local_dir=local_dir)

# Inicializar modelos
print("Inicializando modelos...")
with torch.inference_mode():
    # Initialize nodes
    intconstant = NODE_CLASS_MAPPINGS["INTConstant"]()
    dualcliploader = NODE_CLASS_MAPPINGS["DualCLIPLoader"]()
    dualcliploader_357 = dualcliploader.load_clip(
        clip_name1="models/text_encoders/t5xxl_fp16.safetensors",
        clip_name2="models/text_encoders/ViT-L-14-TEXT-detail-improved-hiT-GmP-HF.safetensors",
        type="flux",
    )
    stylemodelloader = NODE_CLASS_MAPPINGS["StyleModelLoader"]()
    stylemodelloader_441 = stylemodelloader.load_style_model(
        style_model_name="models/style_models/flux1-redux-dev.safetensors"
    )
    vaeloader = NODE_CLASS_MAPPINGS["VAELoader"]()
    vaeloader_359 = vaeloader.load_vae(vae_name="models/vae/ae.safetensors")

    # Carregar modelos na GPU
    model_loaders = [dualcliploader_357, vaeloader_359, stylemodelloader_441]
    valid_models = [
        getattr(loader[0], 'patcher', loader[0]) 
        for loader in model_loaders
        if not isinstance(loader[0], dict) and not isinstance(getattr(loader[0], 'patcher', None), dict)
    ]
    model_management.load_models_gpu(valid_models)

def import_custom_nodes():
    import asyncio
    import execution
    from nodes import init_extra_nodes
    import server

    loop = asyncio.new_event_loop()
    asyncio.set_event_loop(loop)
    server_instance = server.PromptServer(loop)
    execution.PromptQueue(server_instance)
    init_extra_nodes()

@spaces.GPU
def generate_image(prompt, input_image, lora_weight, progress=gr.Progress(track_tqdm=True)):
    """Função principal de geração com monitoramento de progresso"""
    import_custom_nodes()
    try:
        with torch.inference_mode():
            # Codificar texto
            cliptextencode = NODE_CLASS_MAPPINGS["CLIPTextEncode"]()
            encoded_text = cliptextencode.encode(
                text=prompt,
                clip=get_value_at_index(dualcliploader_357, 0)
            )

            # Carregar LoRA
            loraloadermodelonly = NODE_CLASS_MAPPINGS["LoraLoaderModelOnly"]()
            lora_model = loraloadermodelonly.load_lora_model_only(
                lora_name="models/lora/NFTNIK_FLUX.1[dev]_LoRA.safetensors",
                strength_model=lora_weight,
                model=get_value_at_index(stylemodelloader_441, 0)
            )

            # Processar imagem
            loadimage = NODE_CLASS_MAPPINGS["LoadImage"]()
            loaded_image = loadimage.load_image(image=input_image)

            # Decodificar
            vaedecode = NODE_CLASS_MAPPINGS["VAEDecode"]()
            decoded = vaedecode.decode(
                samples=get_value_at_index(lora_model, 0),
                vae=get_value_at_index(vaeloader_359, 0)
            )

            # Salvar imagem
            temp_filename = f"Flux_{random.randint(0, 99999)}.png"
            temp_path = os.path.join(output_dir, temp_filename)
            Image.fromarray((get_value_at_index(decoded, 0) * 255).astype("uint8")).save(temp_path)

            return temp_path
    except Exception as e:
        print(f"Erro ao gerar imagem: {str(e)}")
        return None

# Interface Gradio
with gr.Blocks() as app:
    gr.Markdown("# Gerador de Imagens FLUX Redux")
    with gr.Row():
        with gr.Column():
            prompt_input = gr.Textbox(label="Prompt", placeholder="Digite seu prompt aqui...", lines=5)
            input_image = gr.Image(label="Imagem de Entrada", type="filepath")
            lora_weight = gr.Slider(minimum=0, maximum=2, step=0.1, value=0.6, label="Peso LoRA")
            generate_btn = gr.Button("Gerar Imagem")

        with gr.Column():
            output_image = gr.Image(label="Imagem Gerada", type="filepath")

    generate_btn.click(
        fn=generate_image,
        inputs=[prompt_input, input_image, lora_weight],
        outputs=[output_image]
    )

if __name__ == "__main__":
    # Download models at startup
    download_models()
    # Launch the app
    app.launch()