app.py

import gradio as gr
from transformers import Qwen2VLForConditionalGeneration, AutoTokenizer, AutoProcessor
from qwen_vl_utils import process_vision_info

# Load the model and processor
model = Qwen2VLForConditionalGeneration.from_pretrained(
    "Qwen/Qwen2-VL-72B-Instruct", torch_dtype="auto", device_map="auto"
)
processor = AutoProcessor.from_pretrained("Qwen/Qwen2-VL-7B-Instruct")

# Define a function to process input and generate a response
def generate_response(image, text):
    # Prepare the input
    messages = [
        {
            "role": "user",
            "content": [
                {"type": "image", "image": image},
                {"type": "text", "text": text},
            ],
        }
    ]
    
    # Process the input data
    text_data = processor.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
    image_inputs, video_inputs = process_vision_info(messages)
    inputs = processor(
        text=[text_data],
        images=image_inputs,
        videos=video_inputs,
        padding=True,
        return_tensors="pt",
    )

    # Generate the output
    generated_ids = model.generate(**inputs, max_new_tokens=128)
    generated_ids_trimmed = [
        out_ids[len(in_ids):] for in_ids, out_ids in zip(inputs.input_ids, generated_ids)
    ]
    output_text = processor.batch_decode(
        generated_ids_trimmed, skip_special_tokens=True, clean_up_tokenization_spaces=False
    )
    
    return output_text[0]

# Create the Gradio interface
interface = gr.Interface(
    fn=generate_response,
    inputs=[gr.Image(type="pil", label="Input Image"), gr.Textbox(label="Input Text")],
    outputs="text",
    title="Qwen2-VL-72B-Instruct",
    description="Generate AI responses based on image and text input using Qwen2-VL-72B-Instruct.",
)

# Launch the app
interface.launch()