Luma

Text to video

Turn a text prompt into an AI video

Text-to-video generation creates footage from a written scene description. Describe the subject, action, environment and camera direction, choose from the models and settings currently shown, and review the exact credit cost before creating. No source footage is required, and completed videos are saved to your private library.

Available models, settings and costs are shown inside the product before you create.

Start withA written prompt
Prompt

A cyclist crosses a quiet city street as the camera tracks alongside

ModelSettingsCreate · exact credits
Saved to your libraryGenerated scene

How it works

From input to video in three steps

  1. 01

    Write the scene

    Describe what is visible, what happens and how the camera should observe it. A concrete prompt is easier to interpret.

  2. 02

    Choose available settings

    Select from the current models and controls shown for text-to-video generation in the composer.

  3. 03

    Review and create

    Check the exact credit cost on the action button, submit the request and find the completed result in your library.

Practical guidance

Write prompts that are easy to picture

A good prompt reads like a compact shot description. Start with the important visual facts, then add motion and camera direction.

Name the subject and action

Say who or what is in the scene and what it is doing. Put the main action before decorative details.

Add place and atmosphere

Mention the environment, time of day or lighting when those details matter to the shot.

Give the camera one direction

Use a clear instruction such as a slow push, a wide static shot or a tracking camera when movement matters.

Questions

Frequently asked questions

Do I need an image or video to start?

No. Text-to-video generation starts with a written prompt and does not require source media.

What should I include in a text-to-video prompt?

Include the main subject, the action, the environment and any important camera direction. Keep the central idea clear before adding detail.

Which settings can I control?

The composer shows the current models and settings available for text-to-video generation. Controls can differ by model.

When do I see the credit cost?

The create button displays the exact credit cost after you choose the available model and settings, before you submit the generation.

Start with the idea you have now

Write the scene or add your source image. You will see the available controls and exact credit cost before creating.

Create a Video