# Multimodal Model

> A model that can process or generate multiple kinds of data, such as text, images, audio, video, or code.

- Category: Models & Training
- Stage: Established
- Also called: Omnimodal model, Vision-language model
- Canonical: https://www.joelcsaji.com/ai-dictionary/multimodal-model

## Plain English

It can understand more than words.

## Example

A model reads a screenshot, explains the UI bug, and suggests CSS changes.

## Why it matters

Multimodal capability makes agents useful in browsers, design tools, documents, and physical-world workflows.

## Related

- [Computer Use](https://www.joelcsaji.com/ai-dictionary/computer-use)
- [Grounding](https://www.joelcsaji.com/ai-dictionary/grounding)
- [Frontier Model](https://www.joelcsaji.com/ai-dictionary/frontier-model)

## Sources

- [Google Cloud Generative AI Glossary](https://docs.cloud.google.com/docs/generative-ai/glossary)
- [Stanford HAI: 2026 AI Index](https://hai.stanford.edu/ai-index/2026-ai-index-report)