app.py

import gradio as gr
import spaces
import torch
from diffusers import Transformer2DModel
from scripts.diffusers_patches import pixart_sigma_init_patched_inputs, PixArtSigmaPipeline

assert getattr(Transformer2DModel, '_init_patched_inputs', False), "Need to Upgrade diffusers: pip install git+https://github.com/huggingface/diffusers"
setattr(Transformer2DModel, '_init_patched_inputs', pixart_sigma_init_patched_inputs)
device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
weight_dtype = torch.float16

transformer = Transformer2DModel.from_pretrained(
    "PixArt-alpha/PixArt-Sigma-XL-2-1024-MS",
    subfolder='transformer',
    torch_dtype=weight_dtype,
    use_safetensors=True,
)
pipe = PixArtSigmaPipeline.from_pretrained(
    "PixArt-alpha/pixart_sigma_sdxlvae_T5_diffusers",
    transformer=transformer,
    torch_dtype=weight_dtype,
    use_safetensors=True,
)
pipe.to(device)

@spaces.GPU(duration=120)
def generate(prompt, negative_prompt, num_inference_steps, guidance_scale, height, width):
    image = pipe(
        prompt,
        negative_prompt=negative_prompt,
        num_inference_steps=num_inference_steps,
        guidance_scale=guidance_scale,
        height=height,
        width=width
    ).images[0]
    return image

interface = gr.Interface(
    fn=generate,
    inputs=[
        gr.Text(label="Prompt"),
        gr.Text(label="Negative Prompt"),
        gr.Slider(minimum=1, maximum=500, value=100, step=1, label="Number of Inference Steps"),
        gr.Slider(minimum=1, maximum=20, value=4.5, step=0.1, label="Guidance Scale"),
        gr.Slider(minimum=64, maximum=1024, value=512, step=64, label="Height"),
        gr.Slider(minimum=64, maximum=1024, value=512, step=64, label="Width"),
    ],
    outputs=gr.Image(label="Generated Image"),
    title="PixArt Sigma Image Generation",
    description="Generate images using the PixArt Sigma model.",
)

interface.launch()