Engage in multi-modal conversations with images and videos
Extract text from images in multiple languages