import gradio as gr
from transformers import AutoModelForCausalLM, AutoTokenizer, AutoProcessor
from PIL import Image
import torch

# Load Qwen-VL model
model_id = "Qwen/Qwen-VL-Chat"
processor = AutoProcessor.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(model_id, device_map="auto", trust_remote_code=True).eval()
tokenizer = AutoTokenizer.from_pretrained(model_id, trust_remote_code=True)

def recognize_handwriting(img):
    prompt = "<|im_start|>user\nWhat is written in the image?\n<|im_end|>\n<|im_start|>assistant\n"
    inputs = processor(text=prompt, images=img, return_tensors="pt").to(model.device)
    outputs = model.generate(**inputs, max_new_tokens=128)
    result = processor.batch_decode(outputs, skip_special_tokens=True)[0]
    return result

gr.Interface(fn=recognize_handwriting,
             inputs=gr.Image(type="pil"),
             outputs="text",
             title="Handwriting Recognition with Qwen-VL").launch()