A multimodal model is a type of artificial intelligence system in which multiple input sources are used to generate a single output.
"The document-reading tool's multimodal model allows it to process all elements of a file (visual, text, video and audio) to determine what the document is and what each component means. It then takes action based on that analysis."
