What Multimodal Actually Means in Claude (Stop Typing, Start Showing)

Everyone keeps calling Claude "multimodal" like you're supposed to know what that means. It's actually dead simple.

Multimodal just means Claude can process more than text. You can feed it images, screenshots, PDFs, charts, spreadsheets, code files. It sees all of it. (The word sounds like a printer setting. The capability is anything but.)

Here's the 30-second version:

Stop Describing. Start Showing.

Instead of describing a bug to Claude, screenshot the error and drop it in. Instead of typing out a table, upload the CSV. Instead of explaining what a design looks like, send the image.

Every minute you spend writing a paragraph that describes something visible on your screen is a minute you donated to nobody.

The One Exception

The one thing Claude can't process right now is video. Everything else is fair game. Once you start thinking in multimodal, you stop typing long descriptions of things entirely and just show Claude what you're looking at.

If you can see it, Claude can see it. Stop translating.

This is part 6 of the 20-part Cowork series. All 20 concepts are in the full guide.

Next time you start typing "so there's this error that says...", stop. Screenshot. Drop. Done.