A media company wants to build an internal tool that can accept a customer's uploaded product photo, a voice recording describing a defect, and a typed message, then generate a single unified support summary. Which type of generative AI model is best suited to natively handle all of these input types together?