AI Glossary

Multimodal AI

Also called multimodal model

What does “Multimodal AI” mean?

Multimodal AI: AI that can take in or produce more than one type of data, such as text, images, audio and video, within the same model. Rather than separate tools for each, a single multimodal model can look at a photo and answer questions about it in words, transcribe speech or generate an image from a description.

Multimodal AI visual guide: AI that can take in or produce more than one type of data, such as text, images, audio and video, within the same model. Rather than separate tools for each, a single multimodal model can look at a photo and answer questions about it in words, transcribe speech or generate an image from a description.
Visual guide Open full size (new tab)

What is an example of “Multimodal AI”?

Asking an assistant to read a photographed receipt and total the items requires a model that understands both the image and your text request.

Keep learning

See also

Vision-language modelTermGenerative AITermComputer visionTermFoundation modelTerm

Same topic

Speech-to-textTermNarrow AITermModelTermNatural language processingTermMachine learningTerm