app.py

import torch
import gradio as gr
from transformers import AutoProcessor, AutoModelForImageTextToText

# Load Model and Processor
MODEL_NAME = "Qwen/Qwen2.5-VL-72B-Instruct"
processor = AutoProcessor.from_pretrained(MODEL_NAME)
model = AutoModelForImageTextToText.from_pretrained(MODEL_NAME, torch_dtype=torch.float16, device_map="auto")

# Function to handle text inputs
def chatbot(message):
    inputs = processor([{"role": "user", "content": message}], return_tensors="pt")
    inputs = {k: v.to(model.device) for k, v in inputs.items()}  # Move inputs to the correct device
    outputs = model.generate(**inputs, max_length=512)
    response = processor.batch_decode(outputs, skip_special_tokens=True)[0]
    return response

# Gradio UI
with gr.Blocks() as demo:
    gr.Markdown("# 🤖 AI Chatbot using Qwen 2.5-VL-72B-Instruct")
    chat = gr.ChatInterface(fn=chatbot)

# Launch Gradio App
demo.launch()