import torch
from diffusers import DiffusionPipeline
import gradio as gr
import os
import spaces

model_list = [model.strip() for model in os.environ.get("MODELS").split(",")]
lora_list = [model.strip() for model in os.environ.get("LORAS").split(",")]


models = {}
for model_name in model_list:
    try:
        models[model_name] = DiffusionPipeline.from_pretrained(model_name).to("cuda")
    except Exception as e:
        print(f"Error loading model {model_name}: {e}")


@spaces.GPU
def generate_image(model_name, prompt, negative_prompt, num_inference_steps, guidance_scale):
    pipe = models[model_name]
    output = pipe(prompt, negative_prompt=negative_prompt, num_inference_steps=num_inference_steps, guidance_scale=guidance_scale)["images"][0]
    return output

    

# Create the Gradio blocks
with gr.Blocks() as demo:
    with gr.Row():
        with gr.Column():
            model_dropdown = gr.Dropdown(choices=list(models.keys()), value=model_list[0] if model_list else None, label="Model")
            prompt = gr.Textbox(label="Prompt")
            negative_prompt = gr.Textbox(label="Negative Prompt", value="")
            num_inference_steps = gr.Slider(minimum=10, maximum=50, step=1, value=25, label="Number of Inference Steps")
            guidance_scale = gr.Slider(minimum=1, maximum=20, step=0.5, value=7.5, label="Guidance Scale")
        with gr.Column():
            output_image = gr.Image(label="Generated Image")
    generate_btn = gr.Button("Generate Image")
    generate_btn.click(generate_image, inputs=[model_dropdown, prompt, negative_prompt, num_inference_steps, guidance_scale], outputs=output_image)

demo.launch()