app.py

import gradio as gr
import numpy as np
import random
from huggingface_hub import hf_hub_download

import spaces  # [uncomment to use ZeroGPU]
from diffusers import FluxPipeline
import torch

device = "cuda" if torch.cuda.is_available() else "cpu"
model_repo_id = "black-forest-labs/FLUX.1-schnell"  # Replace to the model you would like to use
torch_dtype = torch.bfloat16

# pipe = FluxPipeline.from_pretrained(model_repo_id, torch_dtype=torch_dtype)
# pipe = pipe.to(device)

# load pruned model
pruned_pipe = FluxPipeline.from_pretrained(model_repo_id, torch_dtype=torch_dtype)
pruned_pipe.transformer = torch.load(
    hf_hub_download("zhangyang-0123/EcoDiffPrunedModels", "model/flux/flux.pkl"),
    map_location="cpu",
)
pruned_pipe = pruned_pipe.to(device)


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


@spaces.GPU
def generate_images(prompt, seed, steps):
    # Run the model and return images directly
    # g_cpu = torch.Generator("cuda").manual_seed(seed)
    # original_image = pipe(prompt=prompt, generator=g_cpu, num_inference_steps=steps).images[0]
    g_cpu = torch.Generator("cuda").manual_seed(seed)
    ecodiff_image = pruned_pipe(prompt=prompt, generator=g_cpu, num_inference_steps=steps).images[0]
    return ecodiff_image


examples = [
    "A clock tower floating in a sea of clouds",
    "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k",
    "An astronaut riding a green horse",
    "A delicious ceviche cheesecake slice",
    "A sprawling cyberpunk metropolis at night, with towering skyscrapers emitting neon lights of every color, holographic billboards advertising alien languages",
]

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

header = """
# 🌱 EcoDiff Pruned FLUX-Schnell (20% Pruning Ratio)

We are not able to host two FLUX models in the same space, so we only show the pruned model here. **👉 [Click here to compare with the Original FLUX Model](https://huggingface.co/spaces/black-forest-labs/FLUX.1-schnell)**.
"""

header_2 = """
<div style="text-align: center; display: flex; justify-content: left; gap: 5px;">
    <a href="https://arxiv.org/abs/2412.02852"><img src="https://img.shields.io/badge/ariXv-Paper-A42C25.svg" alt="arXiv"></a>
    <a href="https://huggingface.co/zhangyang-0123/EcoDiffPrunedModels"><img src="https://img.shields.io/badge/🤗-Model-ffbd45.svg" alt="HuggingFace"></a>
</div>
"""

with gr.Blocks() as demo:
    gr.Markdown(header)
    gr.HTML(header_2)
    with gr.Row():
        prompt = gr.Textbox(
            label="Prompt",
            value="A clock tower floating in a sea of clouds",
            scale=3,
        )
        seed = gr.Number(label="Seed", value=44, precision=0, scale=1)
        steps = gr.Slider(
            label="Number of Steps",
            minimum=1,
            maximum=100,
            value=5,
            step=1,
            scale=1,
        )
        generate_btn = gr.Button("Generate Images")
    gr.Examples(
        examples=[
            "A clock tower floating in a sea of clouds",
            "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k",
            "An astronaut riding a green horse",
            "A delicious ceviche cheesecake slice",
            "A sprawling cyberpunk metropolis at night, with towering skyscrapers emitting neon lights of every color, holographic billboards advertising alien languages",
        ],
        inputs=[prompt],
    )
    with gr.Row():
        # original_output = gr.Image(label="Original Output")
        ecodiff_output = gr.Image(label="EcoDiff Output")
    gr.on(
        triggers=[generate_btn.click, prompt.submit],
        fn=generate_images,
        inputs=[
            prompt,
            seed,
            steps,
        ],
        outputs=[ecodiff_output],
    )

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