Video models fail differently from image models. A still frame rewards description, while a clip rewards verbs: what moves, how the camera moves, and how long the moment is. A prompt that reads like an image caption tends to come back as a nearly static shot with drifting detail.
This page starts in video mode, so the controls in front of you describe action and camera behaviour plus duration and aspect ratio. The result is a short prompt you can send on its own, or attach to a start image — in which case leave the frame description out, because the model already has the frame.
Build a prompt
Nothing here talks to a server: the wording is assembled in your browser from the choices below, so it costs nothing and works with an ad blocker on.
Shot, motion and look, written the way video models read prompts.
Audio
Some video models generate sound with the clip. If yours does not, plan to add the track in an editor instead of asking for it in the prompt.
Video prompt
a single continuous take of a slow reveal of the subject, slow dolly in, soft directional window light, cinematic filmic grade with gentle contrast, steady, coherent motion across the whole clip.
Video models read motion verbs best. Describe what moves and how the camera behaves; do not re-describe a frame you have already uploaded as a start image.
Settings to pair with it
Duration: 5 seconds Aspect ratio: 16:9 Audio: silent — add the track in your editor
Duration and aspect are the two settings people forget, then wonder why the clip crops badly.
No account, no API key, no generation. This page only writes the text — you paste it into whichever tool you already pay for. Nothing you type here is uploaded, and this site is not affiliated with any generation service.
How to use
- Describe your subject as a noun phrase, not a sentence: "a single ceramic coffee cup", not "I want a picture of a cup".
- Choose a setting and a light. Light changes more about the result than any adjective you can add.
- Pick a lens or framing and a medium, so the model knows whether it is shooting a photograph or painting an illustration.
- Add a palette and mood to lock the colour range and the feeling.
- Copy the prompt, send it with the aspect ratio the tool reports, then change exactly one word between runs so you learn what each word does.
Frequently asked questions
Should I describe the first frame if I am uploading one?
No. Re-describing an uploaded frame wastes the prompt and can make the model re-render it. Use the text for motion instead: what the subject does and how the camera behaves.
How long should the prompt be?
Shorter than an image prompt, usually one sentence plus a phrase about camera behaviour. Duration and aspect ratio belong in the settings fields, not in the text.
What actually makes a clip look wrong?
Two or three competing motions. Asking for an orbiting camera, a walking subject and falling rain in a five-second clip is asking for the model to drop one of them mid-shot.