app.py

import gradio as gr
import numpy as np
import random

# import spaces #[uncomment to use ZeroGPU]
from diffusers import DiffusionPipeline, StableDiffusionPipeline
from peft import PeftModel, LoraConfig
import torch
from rembg import remove

model_repo_id = "stabilityai/sdxl-turbo"  # Replace to the model you would like to use

MAX_SEED = np.iinfo(np.int32).max
MAX_IMAGE_SIZE = 1024



# @spaces.GPU #[uncomment to use ZeroGPU]
def infer(
    model,
    prompt,
    negative_prompt,
    seed,
    randomize_seed,
    width,
    height,
    guidance_scale,
    lora_scale,
    num_inference_steps,
    remove_background,
    progress=gr.Progress(track_tqdm=True),
):
    if randomize_seed:
        seed = random.randint(0, MAX_SEED)

    generator = torch.Generator().manual_seed(seed)

    device = "cuda" if torch.cuda.is_available() else "cpu"

    if torch.cuda.is_available():
        torch_dtype = torch.float16
    else:
        torch_dtype = torch.float32

    if model == "Ramzes":
        pipe = StableDiffusionPipeline.from_pretrained("stable-diffusion-v1-5/stable-diffusion-v1-5", torch_dtype=torch_dtype)
        pipe.unet = PeftModel.from_pretrained(pipe.unet, "Bordoglor/Ramzes_adapter_sd_v1.5", subfolder="unet")
        pipe.text_encoder = PeftModel.from_pretrained(pipe.text_encoder, "Bordoglor/Ramzes_adapter_sd_v1.5", subfolder="text_encoder")
        pipe.unet.load_state_dict({k: lora_scale*v for k, v in pipe.unet.state_dict().items()})
        pipe.text_encoder.load_state_dict({k: lora_scale*v for k, v in pipe.text_encoder.state_dict().items()})
    else:     
        pipe = DiffusionPipeline.from_pretrained(model, torch_dtype=torch_dtype)
    pipe = pipe.to(device)

    image = pipe(
        prompt=prompt,
        negative_prompt=negative_prompt,
        guidance_scale=guidance_scale,
        num_inference_steps=num_inference_steps,
        width=width,
        height=height,
        generator=generator,
    ).images[0]

    if remove_background:
        image = remove(image)

    return image, seed

css = """
#col-container {
    margin: 0 auto;
    max-width: 640px;
}
"""

with gr.Blocks(css=css) as demo:
    with gr.Column(elem_id="col-container"):
        gr.Markdown(" # Text-to-Image Gradio Template")


        with gr.Row():
            model = gr.Dropdown(
                choices=["stabilityai/sdxl-turbo", "CompVis/stable-diffusion-v1-4", "stable-diffusion-v1-5/stable-diffusion-v1-5", "Ramzes"],
                value=model_repo_id,
                label="Model",
                info="Choose which diffusion model to use"
            )

        with gr.Row():
                remove_background = gr.Checkbox(
                    label="Delete background?", value=True
                )
            
        with gr.Row():
            prompt = gr.Text(
                label="Prompt",
                show_label=False,
                max_lines=1,
                placeholder="Enter your prompt",
                container=False,
            )

            run_button = gr.Button("Run", scale=0, variant="primary")

        result = gr.Image(label="Result", show_label=False)

        with gr.Accordion("Advanced Settings", open=True):
            negative_prompt = gr.Text(
                label="Negative prompt",
                max_lines=1,
                placeholder="Enter a negative prompt",
                visible=True,
            )

            seed = gr.Slider(
                label="Seed",
                minimum=0,
                maximum=MAX_SEED,
                step=1,
                value=42,
            )

            randomize_seed = gr.Checkbox(label="Randomize seed", value=True)
                
            with gr.Row():
                width = gr.Slider(
                    label="Width",
                    minimum=256,
                    maximum=MAX_IMAGE_SIZE,
                    step=32,
                    value=1024,  # Replace with defaults that work for your model
                )

                height = gr.Slider(
                    label="Height",
                    minimum=256,
                    maximum=MAX_IMAGE_SIZE,
                    step=32,
                    value=1024,  # Replace with defaults that work for your model
                )

            with gr.Row():
                guidance_scale = gr.Slider(
                    label="Guidance scale",
                    minimum=0.0,
                    maximum=10.0,
                    step=0.1,
                    value=7.0,  # Replace with defaults that work for your model
                )

                lora_scale = gr.Slider(
                    label="LoRA scale",
                    minimum=0.0,
                    maximum=1.0,
                    step=0.05,
                    value=0.9
                )

                num_inference_steps = gr.Slider(
                    label="Number of inference steps",
                    minimum=1,
                    maximum=50,
                    step=1,
                    value=3,  # Replace with defaults that work for your model
                )

    gr.on(
        triggers=[run_button.click, prompt.submit],
        fn=infer,
        inputs=[
            model,
            prompt,
            negative_prompt,
            seed,
            randomize_seed,
            width,
            height,
            guidance_scale,
            lora_scale,
            num_inference_steps,
            remove_background
        ],
        outputs=[result, seed],
    )

if __name__ == "__main__":
    demo.launch()