app.py

import gradio as gr
import numpy as np
import spaces
import torch
import random
from PIL import Image

from diffusers import FluxKontextPipeline
from diffusers.utils import load_image

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

pipe = FluxKontextPipeline.from_pretrained(
    "black-forest-labs/FLUX.1-Kontext-dev", torch_dtype=torch.bfloat16
).to("cuda")


@spaces.GPU
def infer(
    input_image,
    prompt,
    seed=42,
    randomize_seed=False,
    guidance_scale=2.5,
    steps=28,
    progress=gr.Progress(track_tqdm=True),
):
    """
    Edit an image using AI based on text instructions.

    Args:
        input_image (optional): Path to the image file to edit (if None, generates from text only)
        prompt (required): Text describing what to change (e.g. "remove glasses", "add a hat", "change background to beach")
        seed (optional): Random seed for reproducibility (default: 42)
        randomize_seed (optional): Use random seed instead of fixed seed (default: False)
        guidance_scale (optional): How closely to follow the prompt, 1.0-10.0 (default: 2.5)
        steps (optional): Number of generation steps, 1-30 (default: 28)
        progress (optional): Gradio progress tracker (automatically provided)

    Returns:
        tuple: (edited_image, seed_used, gradio_update)

    Example:
        infer(input_image="/path/to/photo.jpg", prompt="Add sunglasses")
    """
    if randomize_seed:
        seed = random.randint(0, MAX_SEED)

    if input_image:
        input_image = input_image.convert("RGB")
        image = pipe(
            image=input_image,
            prompt=prompt,
            guidance_scale=guidance_scale,
            num_inference_steps=steps,
            generator=torch.Generator().manual_seed(seed),
        ).images[0]
    else:
        image = pipe(
            prompt=prompt,
            guidance_scale=guidance_scale,
            num_inference_steps=steps,
            generator=torch.Generator().manual_seed(seed),
        ).images[0]
    return image, seed, gr.update(visible=True)


css = """
#col-container {
    margin: 0 auto;
    max-width: 960px;
}
"""

with gr.Blocks(css=css) as demo:

    with gr.Column(elem_id="col-container"):
        gr.Markdown(
            f"""# FLUX.1 Kontext [dev]
Image editing and manipulation model guidance-distilled from FLUX.1 Kontext [pro], [[blog]](https://bfl.ai/announcements/flux-1-kontext-dev) [[model]](https://huggingface.co/black-forest-labs/FLUX.1-Kontext-dev)
        """
        )
        with gr.Row():
            with gr.Column():
                input_image = gr.Image(label="Upload the image for editing", type="pil")
                with gr.Row():
                    prompt = gr.Text(
                        label="Prompt",
                        show_label=False,
                        max_lines=1,
                        placeholder="Enter your prompt for editing (e.g., 'Remove glasses', 'Add a hat')",
                        container=False,
                    )
                    run_button = gr.Button("Run", scale=0)
                with gr.Accordion("Advanced Settings", open=False):

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

                    randomize_seed = gr.Checkbox(label="Randomize seed", value=True)

                    guidance_scale = gr.Slider(
                        label="Guidance Scale",
                        minimum=1,
                        maximum=10,
                        step=0.1,
                        value=2.5,
                    )

                    steps = gr.Slider(
                        label="Steps", minimum=1, maximum=30, value=28, step=1
                    )

            with gr.Column():
                result = gr.Image(label="Result", show_label=False, interactive=False)
                reuse_button = gr.Button("Reuse this image", visible=False)

    gr.on(
        triggers=[run_button.click, prompt.submit],
        fn=infer,
        inputs=[input_image, prompt, seed, randomize_seed, guidance_scale, steps],
        outputs=[result, seed, reuse_button],
    )
    reuse_button.click(fn=lambda image: image, inputs=[result], outputs=[input_image])

demo.launch(mcp_server=True)