app.py

import gradio as gr
from transformers import TrOCRProcessor, VisionEncoderDecoderModel, pipeline
from PIL import Image

# Load OCR model
processor = TrOCRProcessor.from_pretrained('microsoft/trocr-base-stage1')
model = VisionEncoderDecoderModel.from_pretrained('microsoft/trocr-base-stage1')

# Load LLM
llm = pipeline("text-generation", model="distilgpt2")

def process_image(image):
    pil_image = Image.fromarray(image)
    pixel_values = processor(images=pil_image, return_tensors="pt").pixel_values
    generated_ids = model.generate(pixel_values)
    text = processor.batch_decode(generated_ids, skip_special_tokens=True)[0]

    llm_output = llm(text, max_length=100, do_sample=True)[0]["generated_text"]
    return f"OCR Text:\n{text}\n\nLLM Response:\n{llm_output}"

gr.Interface(fn=process_image, 
             inputs=gr.Image(type="numpy"), 
             outputs="text", 
             title="OCR + LLM Text Generator").launch()