L

Learn AI

Track progress · learn offline

Open

Lesson 6 of 8

Image & video

7 min read

You describe a scene in words, and seconds later it's a picture — or a few seconds of video. No design app, no camera. How far does "just describe it" go?

Describe it, and it draws

Ask Gemini for an image — "a cosy reading nook at golden hour, watercolour" — and its image model, Nano Banana, generates it right in the chat. Because it's the same conversation, you refine by talking: "make it evening," "add a cat." You shape a picture the way you shape text — by describing the change.

Image and video generation live inside the chat, so you create and refine by describing — no separate app to learn.

And it moves, too

For motion, Gemini uses Veo — describe a short scene and it produces a few seconds of video, sometimes inside Flow, a tool built for stitching clips into something longer. Same idea as images: you say what you want to see, and it renders it. Great for a quick idea, a mood, a rough cut.

From a sentence, Gemini can hand you a still or a moving clip — the description is the interface.

Generated media gets small details wrong — hands, text on signs, exact logos, physics in a clip. Lovely for ideas and drafts; double-check before anything official.

The shape of it

You want a birthday-card illustration of a cat astronaut — then a two-second clip of it waving. How?

Continue in the app

Take the whole Get the most out of Gemini course — tracked

Get your personalized path, progress and streaks in the app — this lesson and every next one, in order.