app.py

import gradio as gr
import torch
from PIL import Image
from diffusers.utils import load_image
from diffusers.pipelines.flux.pipeline_flux_controlnet import FluxControlNetPipeline
from diffusers.models.controlnet_flux import FluxControlNetModel

base_model = 'black-forest-labs/FLUX.1-dev'
controlnet_model = 'promeai/FLUX.1-controlnet-lineart-promeai'
controlnet = FluxControlNetModel.from_pretrained(controlnet_model, torch_dtype=torch.bfloat16)
pipe = FluxControlNetPipeline.from_pretrained(base_model, controlnet=controlnet, torch_dtype=torch.bfloat16)
pipe.to("cuda")

def generate_image(prompt, control_image, controlnet_conditioning_scale, num_inference_steps, guidance_scale):
    control_image = load_image(control_image) if isinstance(control_image, str) else control_image

    result = pipe(
        prompt,
        control_image=control_image,
        controlnet_conditioning_scale=controlnet_conditioning_scale,
        num_inference_steps=num_inference_steps,
        guidance_scale=guidance_scale,
    ).images[0]
    
    return result

with gr.Blocks() as demo:
    gr.Markdown("# FLUX ControlNet Pipeline Interface")
    
    with gr.Row():
        with gr.Column():
            prompt = gr.Textbox(label="Prompt", lines=3, placeholder="Enter your prompt here...")
            control_image = gr.Image(source="upload", type="filepath", label="Control Image")
            
            controlnet_conditioning_scale = gr.Slider(0.0, 1.0, value=0.6, label="ControlNet Conditioning Scale")
            num_inference_steps = gr.Slider(1, 100, value=28, step=1, label="Number of Inference Steps")
            guidance_scale = gr.Slider(1.0, 10.0, value=3.5, label="Guidance Scale")
            
            generate_button = gr.Button("Generate Image")
        
        with gr.Column():
            output_image = gr.Image(label="Generated Image")
    generate_button.click(
        generate_image, 
        inputs=[prompt, control_image, controlnet_conditioning_scale, num_inference_steps, guidance_scale],
        outputs=output_image
    )

demo.launch()