How to Generate a Short Video with Veo in Gemini

Veo is Google's text-to-video model, available to Gemini subscribers inside the chat app. You describe a scene and it generates a short clip with motion and sound. This guide writes a prompt that produces a usable result and shows how to iterate when the first take is off.

What you need

• A Gemini plan that includes Veo video generation
• A clear visual idea: subject, setting, and motion
• A few minutes per clip for rendering

Step 1: Open the video tool

In the Gemini app, open the tools menu near the prompt box and pick the video option (often labelled Video or Veo). The interface switches to a mode built for generating clips rather than text replies.

Step 2: Write a specific prompt

Good video prompts name the subject, the action, the camera movement, and the style. Vague prompts produce generic footage. Treat it like a one-line shot description a director would hand a camera operator.

A golden retriever puppy running across a sunny beach at sunrise,
slow motion, camera tracking alongside, warm cinematic lighting,
soft waves in the background.

Step 3: Review, then refine

When the clip finishes, watch it and decide what to change. Rather than rewriting from scratch, adjust one element at a time, such as the camera angle or time of day, so you can tell what each change does.

Result

You get a short, downloadable clip from a text description, and a fast loop for refining it. Generating two or three variations and picking the best one is usually quicker than perfecting a single prompt.