app-fast.py

import gradio as gr
import PIL
import spaces
import torch
from diffusers import BitsAndBytesConfig as DiffusersBitsAndBytesConfig
from hi_diffusers import HiDreamImagePipeline, HiDreamImageTransformer2DModel
from hi_diffusers.schedulers.flash_flow_match import (
    FlashFlowMatchEulerDiscreteScheduler,
)
from transformers import (
    AutoModelForCausalLM,
    AutoTokenizer,
)
from transformers import BitsAndBytesConfig as TransformersBitsAndBytesConfig

# Constants
MODEL_PREFIX: str = "HiDream-ai"
LLAMA_MODEL_NAME: str = "meta-llama/Meta-Llama-3.1-8B-Instruct"
MODEL_PATH = "HiDream-ai/HiDream-I1-Fast"
MODEL_CONFIGS = {
    "guidance_scale": 0.0,
    "num_inference_steps": 16,
    "shift": 3.0,
    "scheduler": FlashFlowMatchEulerDiscreteScheduler,
}


# Supported image sizes
RESOLUTION_OPTIONS: list[str] = [
    "1024 x 1024",
    "768 x 1360",
    "1360 x 768",
    "880 x 1168",
    "1168 x 880",
    "1248 x 832",
    "832 x 1248",
]

quant_config = TransformersBitsAndBytesConfig(
    load_in_4bit=True,
)

tokenizer = AutoTokenizer.from_pretrained(LLAMA_MODEL_NAME, use_fast=False)
text_encoder = AutoModelForCausalLM.from_pretrained(
    LLAMA_MODEL_NAME,
    output_hidden_states=True,
    output_attentions=True,
    low_cpu_mem_usage=True,
    quantization_config=quant_config,
    torch_dtype=torch.bfloat16,
)

quant_config = DiffusersBitsAndBytesConfig(
    load_in_4bit=True,
)
transformer = HiDreamImageTransformer2DModel.from_pretrained(
    MODEL_PATH,
    subfolder="transformer",
    quantization_config=quant_config,
    torch_dtype=torch.bfloat16,
)

scheduler = MODEL_CONFIGS["scheduler"](
    num_train_timesteps=1000,
    shift=MODEL_CONFIGS["shift"],
    use_dynamic_shifting=False,
)

pipe = HiDreamImagePipeline.from_pretrained(
    MODEL_PATH,
    transformer=transformer,
    scheduler=scheduler,
    tokenizer_4=tokenizer,
    text_encoder_4=text_encoder,
    device_map="balanced",
    torch_dtype=torch.bfloat16,
)


@spaces.GPU(duration=120)
def generate_image(
    prompt: str,
    resolution: str,
    seed: int,
    progress=gr.Progress(track_tqdm=True),  # noqa: ARG001, B008
) -> tuple[PIL.Image.Image, int]:
    gr.Info(
        "This Spaces is an unofficial quantized version of HiDream-ai-full. It is not as good as the full version, but it is faster and uses less memory."
    )
    if seed == -1:
        seed = torch.randint(0, 1_000_000, (1,)).item()

    height, width = tuple(map(int, resolution.replace(" ", "").split("x")))
    generator = torch.Generator("cuda").manual_seed(seed)

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

    torch.cuda.empty_cache()
    return image, seed


# Gradio UI
with gr.Blocks(title="HiDream Image Generator Fast") as demo:
    gr.Markdown("## 🌈 HiDream Image Generator Fast")

    with gr.Row():
        with gr.Column():
            prompt = gr.Textbox(
                label="Prompt",
                placeholder="e.g. A futuristic city with floating cars at sunset",
                lines=3,
            )

            resolution = gr.Radio(
                choices=RESOLUTION_OPTIONS,
                value=RESOLUTION_OPTIONS[0],
                label="Resolution",
            )

            seed = gr.Number(label="Seed (-1 for random)", value=-1, precision=0)
            generate_btn = gr.Button("Generate Image", variant="primary")
            seed_used = gr.Number(label="Seed Used", interactive=False)

        with gr.Column():
            output_image = gr.Image(label="Generated Image", type="pil")

    generate_btn.click(
        fn=generate_image,
        inputs=[prompt, resolution, seed],
        outputs=[output_image, seed_used],
    )

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