app.py

import gradio as gr
import spaces
from transformers import AutoModelForCausalLM, AutoProcessor
import torch
from PIL import Image
import subprocess
subprocess.run('pip install flash-attn --no-build-isolation', env={'FLASH_ATTENTION_SKIP_CUDA_BUILD': "TRUE"}, shell=True)

models = {
    "microsoft/Phi-3.5-vision-instruct": AutoModelForCausalLM.from_pretrained("microsoft/Phi-3.5-vision-instruct", trust_remote_code=True, torch_dtype="auto", _attn_implementation="flash_attention_2").cuda().eval()
}

processors = {
    "microsoft/Phi-3.5-vision-instruct": AutoProcessor.from_pretrained("microsoft/Phi-3.5-vision-instruct", trust_remote_code=True)
}

default_question = "You are an image to prompt converter. Your work is to observe each and every detail of the image and craft a detailed prompt under 100 words in this format: [image content/subject, description of action, state, and mood], [art form, style], [artist/photographer reference if needed], [additional settings such as camera and lens settings, lighting, colors, effects, texture, background, rendering]."

@spaces.GPU
def run_example(image, text_input=default_question, model_id="microsoft/Phi-3.5-vision-instruct"):
    model = models[model_id]
    processor = processors[model_id]

    prompt = f"<|user|>\n<|image_1|>\n{text_input}<|end|>\n<|assistant|>\n"
    image = Image.fromarray(image).convert("RGB")

    inputs = processor(prompt, image, return_tensors="pt").to("cuda:0")
    generate_ids = model.generate(**inputs, 
                                max_new_tokens=1000,
                                eos_token_id=processor.tokenizer.eos_token_id,
                                )
    generate_ids = generate_ids[:, inputs['input_ids'].shape[1]:]
    response = processor.batch_decode(generate_ids, 
                                    skip_special_tokens=True, 
                                    clean_up_tokenization_spaces=False)[0]
    return response

css = """
  .container {
    border: 2px solid #333;
    padding: 20px;
    max-width: 400px;
    margin: auto;
  }
  #input_img, #output_text {
    border: 2px solid #333;
    width: 100%;
    height: 300px;
    object-fit: cover;
  }
  .gr-button {
    width: 100%;
    margin-top: 10px;
  }
  #copy_button {
    float: right;
    margin-top: -30px;
    cursor: pointer;
  }
"""

with gr.Blocks(css=css) as demo:
    with gr.Box(elem_id="container"):
        input_img = gr.Image(label="Input Picture", elem_id="input_img", type="pil")
        generate_button = gr.Button("Generate Prompt", elem_id="generate_button")
        with gr.Row():
            output_text = gr.Textbox(label="Output Text", elem_id="output_text", interactive=False)
            copy_button = gr.Button("Copy", elem_id="copy_button")

        # Copy functionality
        copy_button.click(fn=lambda text: text, inputs=output_text, outputs=None)

        # Generate functionality
        generate_button.click(run_example, [input_img, default_question], [output_text])

demo.queue(api_open=False)
demo.launch(debug=True, show_api=False)