Model for table extraction that receives an image as input, runs OCR on the image, and returns the text within the image and its bounding boxes.