app.py

import gradio as gr
import numpy as np
import random

import spaces #[uncomment to use ZeroGPU]
from diffusers import DiffusionPipeline
import torch
import subprocess
from groq import Groq
import base64
import os

subprocess.run("rm -rf /data-nvme/zerogpu-offload/*", env={}, shell=True)

# Load FLUX image generator
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
lora_path = "matteomarjanovic/flatsketcher"
weigths_file = "lora.safetensors"

if torch.cuda.is_available():
    torch_dtype = torch.float16
else:
    torch_dtype = torch.float32

pipe = DiffusionPipeline.from_pretrained(model_repo_id, torch_dtype=torch_dtype)
pipe = pipe.to(device)
pipe.load_lora_weights(lora_path, weight_name=weigths_file)

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

def encode_image(image_path):
  with open(image_path, "rb") as image_file:
    return base64.b64encode(image_file.read()).decode('utf-8')


@spaces.GPU #[uncomment to use ZeroGPU]
def infer(
    prompt,
    progress=gr.Progress(track_tqdm=True),
):
    # seed = random.randint(0, MAX_SEED)

    # generator = torch.Generator().manual_seed(seed)

    image = pipe(
        prompt=prompt,
        guidance_scale=0.,
        num_inference_steps=4,
        width=1420,
        height=1080,
        max_sequence_length=256,
        
    ).images[0]

    return image

@spaces.GPU #[uncomment to use ZeroGPU]
def generate_description_fn(
    image,
    progress=gr.Progress(track_tqdm=True),
):
    base64_image = encode_image(image)

    client = Groq(
        api_key=os.environ.get("GROQ_API_KEY"),
    )

    chat_completion = client.chat.completions.create(
        messages=[
            {
                "role": "user",
                "content": [
                    {"type": "text", "text": "What's in this image?"},
                    {
                        "type": "image_url",
                        "image_url": {
                            "url": f"data:image/jpeg;base64,{base64_image}",
                        },
                    },
                ],
            }
        ],
        model="llama-3.2-11b-vision-preview",
    )

    return chat_completion.choices[0].message.content


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

# generated_prompt = ""

with gr.Blocks(css=css) as demo:
    with gr.Row():
        with gr.Column(elem_id="col-input-image"):
            gr.Markdown(" # Drop your image here")
            input_image = gr.Image(type="filepath")
            generate_button = gr.Button("Generate", scale=0, variant="primary")
            generated_prompt = gr.Markdown("")
        with gr.Column(elem_id="col-container"):
            gr.Markdown(" # Text-to-Image Gradio Template")

            with gr.Row():
                prompt = gr.Text(
                    label="Prompt",
                    show_label=False,
                    max_lines=1,
                    placeholder="Enter your prompt",
                    container=False,
                )

                run_button = gr.Button("Run", scale=0, variant="primary")

            result = gr.Image(label="Result", show_label=False)

        gr.on(
            triggers=[run_button.click, prompt.submit],
            fn=infer,
            inputs=[
                prompt,
            ],
            outputs=[result],
        )

        gr.on(
            triggers=[generate_button.click],
            fn=generate_description_fn,
            inputs=[
                input_image,
            ],
            outputs=[generated_prompt],
        )

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