Core Concepts
Multimodal AI
AI that handles text, images, audio, and video in a single model.
Also known as: multimodal,multi-modal AI
Multimodal AI handles multiple types of input and output — text, images, audio, video, code — natively, in a single model. Older AI systems were single-purpose: a text model could not look at a picture; an image model could not read a sentence. Modern frontier models (GPT-5, Claude Opus 4.8, Gemini 3.5 Pro) are multimodal by default — you can paste a screenshot, ask about a chart, upload a PDF with images, or have a real-time voice conversation. The shift matters because it eliminates the friction of switching between specialised tools. Gemini in particular leads on video understanding; OpenAI leads on voice; Claude leads on documents.