app.py

#!/usr/bin/env python

from __future__ import annotations

import os
import random

import gradio as gr
import numpy as np
import PIL.Image
import spaces
import torch
from diffusers import AutoencoderKL, DiffusionPipeline

DESCRIPTION = """
# OpenDalle 1.1

**Demo by [mrfakename](https://mrfake.name/) - [Twitter](https://twitter.com/realmrfakename) - [GitHub](https://github.com/fakerybakery/) - [Hugging Face](https://huggingface.co/mrfakename)**

This is a demo of <a href="https://huggingface.co/dataautogpt3/OpenDalleV1.1">OpenDalle V1.1</a> by @dataautogpt3.

It's a merge of several different models and is supposed to provide excellent performance. Try it out!

[Not Working?](https://huggingface.co/spaces/mrfakename/OpenDalleV1.1-GPU-Demo/discussions/4)

**The code for this demo is based on [@hysts's SD-XL demo](https://huggingface.co/spaces/hysts/SD-XL) running on a A10G GPU.**

**NOTE: The model is licensed under a non-commercial license**

Also see [OpenDalle Original Demo](https://huggingface.co/spaces/mrfakename/OpenDalle-GPU-Demo/)
"""
if not torch.cuda.is_available():
    DESCRIPTION += "\n<h1>Running on CPU 🥶 This demo does not work on CPU. Please use <a href=\"https://huggingface.co/spaces/mrfakename/OpenDalleV1.1-GPU-Demo\">the online demo</a> instead</h1>"

MAX_SEED = np.iinfo(np.int32).max
CACHE_EXAMPLES = torch.cuda.is_available() and os.getenv("CACHE_EXAMPLES", "0") == "1"
MAX_IMAGE_SIZE = int(os.getenv("MAX_IMAGE_SIZE", "1024"))
USE_TORCH_COMPILE = os.getenv("USE_TORCH_COMPILE") == "1"
ENABLE_CPU_OFFLOAD = os.getenv("ENABLE_CPU_OFFLOAD") == "1"
ENABLE_REFINER = os.getenv("ENABLE_REFINER", "0") == "1"

device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
if torch.cuda.is_available():
    vae = AutoencoderKL.from_pretrained("madebyollin/sdxl-vae-fp16-fix", torch_dtype=torch.float16)
    pipe = DiffusionPipeline.from_pretrained(
        "dataautogpt3/ProteusV0.2",
        vae=vae,
        torch_dtype=torch.float16,
        use_safetensors=True,
        variant="fp16",
    )
    if ENABLE_REFINER:
        refiner = DiffusionPipeline.from_pretrained(
            "stabilityai/stable-diffusion-xl-refiner-1.0",
            vae=vae,
            torch_dtype=torch.float16,
            use_safetensors=True,
        )

    if ENABLE_CPU_OFFLOAD:
        pipe.enable_model_cpu_offload()
        if ENABLE_REFINER:
            refiner.enable_model_cpu_offload()
    else:
        pipe.to(device)
        if ENABLE_REFINER:
            refiner.to(device)

    if USE_TORCH_COMPILE:
        pipe.unet = torch.compile(pipe.unet, mode="reduce-overhead", fullgraph=True)
        if ENABLE_REFINER:
            refiner.unet = torch.compile(refiner.unet, mode="reduce-overhead", fullgraph=True)


def randomize_seed_fn(seed: int, randomize_seed: bool) -> int:
    if randomize_seed:
        seed = random.randint(0, MAX_SEED)
    return seed


@spaces.GPU(enable_queue=True)
def generate(
    prompt: str,
    negative_prompt: str = "",
    prompt_2: str = "",
    negative_prompt_2: str = "",
    use_negative_prompt: bool = False,
    use_prompt_2: bool = False,
    use_negative_prompt_2: bool = False,
    seed: int = 0,
    width: int = 1024,
    height: int = 1024,
    guidance_scale_base: float = 5.0,
    guidance_scale_refiner: float = 5.0,
    num_inference_steps_base: int = 25,
    num_inference_steps_refiner: int = 25,
    apply_refiner: bool = False,
    progress=gr.Progress(track_tqdm=True),
) -> PIL.Image.Image:
    print(f"** Generating image for: \"{prompt}\" **")
    generator = torch.Generator().manual_seed(seed)

    if not use_negative_prompt:
        negative_prompt = None  # type: ignore
    if not use_prompt_2:
        prompt_2 = None  # type: ignore
    if not use_negative_prompt_2:
        negative_prompt_2 = None  # type: ignore

    if not apply_refiner:
        return pipe(
            prompt=prompt,
            negative_prompt=negative_prompt,
            prompt_2=prompt_2,
            negative_prompt_2=negative_prompt_2,
            width=width,
            height=height,
            guidance_scale=guidance_scale_base,
            num_inference_steps=num_inference_steps_base,
            generator=generator,
            output_type="pil",
        ).images[0]
    else:
        latents = pipe(
            prompt=prompt,
            negative_prompt=negative_prompt,
            prompt_2=prompt_2,
            negative_prompt_2=negative_prompt_2,
            width=width,
            height=height,
            guidance_scale=guidance_scale_base,
            num_inference_steps=num_inference_steps_base,
            generator=generator,
            output_type="latent",
        ).images
        image = refiner(
            prompt=prompt,
            negative_prompt=negative_prompt,
            prompt_2=prompt_2,
            negative_prompt_2=negative_prompt_2,
            guidance_scale=guidance_scale_refiner,
            num_inference_steps=num_inference_steps_refiner,
            image=latents,
            generator=generator,
        ).images[0]
        return image


examples = [
    "A realistic photograph of an astronaut in a jungle, cold color palette, detailed, 8k",
    "An astronaut riding a green horse",
]

theme = gr.themes.Base(
    font=[gr.themes.GoogleFont('Libre Franklin'), gr.themes.GoogleFont('Public Sans'), 'system-ui', 'sans-serif'],
)
with gr.Blocks(css="footer{display:none !important}", theme=theme) as demo:
    gr.Markdown(DESCRIPTION)
    gr.DuplicateButton(
        value="Duplicate Space for private use",
        elem_id="duplicate-button",
        visible=os.getenv("SHOW_DUPLICATE_BUTTON") == "1",
    )
    with gr.Group():
        prompt = gr.Text(
            label="Prompt",
            show_label=False,
            max_lines=1,
            container=False,
            placeholder="Enter your prompt",
        )
        run_button = gr.Button("Generate")
    result = gr.Image(label="Result", show_label=False)
    with gr.Accordion("Advanced options", open=False):
        with gr.Row():
            use_negative_prompt = gr.Checkbox(label="Use negative prompt", value=False)
            use_prompt_2 = gr.Checkbox(label="Use prompt 2", value=False)
            use_negative_prompt_2 = gr.Checkbox(label="Use negative prompt 2", value=False)
        negative_prompt = gr.Text(
            label="Negative prompt",
            max_lines=1,
            placeholder="Enter a negative prompt",
            visible=False,
        )
        prompt_2 = gr.Text(
            label="Prompt 2",
            max_lines=1,
            placeholder="Enter your prompt",
            visible=False,
        )
        negative_prompt_2 = gr.Text(
            label="Negative prompt 2",
            max_lines=1,
            placeholder="Enter a negative prompt",
            visible=False,
        )

        seed = gr.Slider(
            label="Seed",
            minimum=0,
            maximum=MAX_SEED,
            step=1,
            value=0,
        )
        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,
            )
            height = gr.Slider(
                label="Height",
                minimum=256,
                maximum=MAX_IMAGE_SIZE,
                step=32,
                value=1024,
            )
        apply_refiner = gr.Checkbox(label="Apply refiner", value=False, visible=ENABLE_REFINER)
        with gr.Row():
            guidance_scale_base = gr.Slider(
                label="Guidance scale for base",
                minimum=1,
                maximum=20,
                step=0.1,
                value=5.0,
            )
            num_inference_steps_base = gr.Slider(
                label="Number of inference steps for base",
                minimum=10,
                maximum=100,
                step=1,
                value=25,
            )
        with gr.Row(visible=False) as refiner_params:
            guidance_scale_refiner = gr.Slider(
                label="Guidance scale for refiner",
                minimum=1,
                maximum=20,
                step=0.1,
                value=5.0,
            )
            num_inference_steps_refiner = gr.Slider(
                label="Number of inference steps for refiner",
                minimum=10,
                maximum=100,
                step=1,
                value=25,
            )

    gr.Examples(
        examples=examples,
        inputs=prompt,
        outputs=result,
        fn=generate,
        cache_examples=CACHE_EXAMPLES,
    )

    use_negative_prompt.change(
        fn=lambda x: gr.update(visible=x),
        inputs=use_negative_prompt,
        outputs=negative_prompt,
        queue=False,
        api_name=False,
    )
    use_prompt_2.change(
        fn=lambda x: gr.update(visible=x),
        inputs=use_prompt_2,
        outputs=prompt_2,
        queue=False,
        api_name=False,
    )
    use_negative_prompt_2.change(
        fn=lambda x: gr.update(visible=x),
        inputs=use_negative_prompt_2,
        outputs=negative_prompt_2,
        queue=False,
        api_name=False,
    )
    apply_refiner.change(
        fn=lambda x: gr.update(visible=x),
        inputs=apply_refiner,
        outputs=refiner_params,
        queue=False,
        api_name=False,
    )

    gr.on(
        triggers=[
            prompt.submit,
            negative_prompt.submit,
            prompt_2.submit,
            negative_prompt_2.submit,
            run_button.click,
        ],
        fn=randomize_seed_fn,
        inputs=[seed, randomize_seed],
        outputs=seed,
        queue=False,
        api_name=False,
    ).then(
        fn=generate,
        inputs=[
            prompt,
            negative_prompt,
            prompt_2,
            negative_prompt_2,
            use_negative_prompt,
            use_prompt_2,
            use_negative_prompt_2,
            seed,
            width,
            height,
            guidance_scale_base,
            guidance_scale_refiner,
            num_inference_steps_base,
            num_inference_steps_refiner,
            apply_refiner,
        ],
        outputs=result,
        api_name="run",
    )

if __name__ == "__main__":
    demo.queue(max_size=20, api_open=False).launch(show_api=False)