What are AI models that handle several input types at once called?

A multimodal model is an AI system built to process and understand more than one type of input, such as text, images, audio, and video, within a single unified system, rather than being limited to written language the way earlier chatbots were. This lets you show it a photograph and ask about it, play it an audio clip, or share a chart, and have it respond meaningfully by drawing connections across those different kinds of information.

Open-weight models are AI models whose internal parameters are publicly released for anyone to download and use, a different concept concerning access and licensing rather than what input types a model can process. Offline models simply run without needing an internet connection, another unrelated characteristic having nothing to do with handling multiple formats.

This blending of input types is what lets modern AI assistants describe what's happening in a photo, summarize a chart, or answer spoken questions about a video, capabilities that would have required entirely separate, specialized systems just a few years earlier.