Definition: multimodal AI

An AI model that supports two or more forms of input; for example, text and images. Various versions of ChatGPT and Gemini are trained on text, images, audio and video. See GPT and multimodal.