Text-to-Video Dance Prompts for Wipe Me Down Clips

The text-to-video mode in the studio takes a written description and returns a Wipe Me Down dance clip without an uploaded photo. That means you invent the subject, the location, the light and the camera position from plain words. This guide breaks the prompt into five parts, gives you twelve copy-paste prompts sorted by the four studio looks, and shows you how to change one clause at a time instead of rewriting from nothing.

Before you spend credits, you can test one prompt with the free preview render: 480p, 5 seconds, one per account while the daily preview budget lasts. That is enough to check whether the subject and motion read clearly. The standard render is 720p, 12 seconds, 26 credits. A 480p test render costs 12 credits, and a 1080p render costs 64 credits. The studio also offers 5, 8 and 10 second lengths, and 9:16, 16:9 or 1:1 frames.

Prompt structure for a photo-free dance clip

Five clauses in this order.

| Clause | What it does | Example | | Subject | Who appears, described from scratch | one dancer in a grey hoodie and jeans | | Setting | Where the dancer stands | an empty parking garage at night | | Lighting | One light direction and quality | a single overhead fluorescent tube | | Camera | One framing or move | camera locked off at waist height | | Motion | One dance action per prompt | steps side to side on the beat |

Keep each clause to a few words. Do not list more than one subject, one location or one camera move. A single clear motion reads better than a list of three movements.

A template:

a [subject description] [motion phrase] in [setting], [lighting], camera [locked off / pushes in slowly / drifts left], [length] seconds, [frame]

Example filled:

a young man in a black hoodie and loose jeans steps side to side on the beat in an empty concrete room, one hard light from camera left, camera locked off at chest height, 12 seconds, 9:16

The length and frame can be set with the sliders instead of typed, but including them in the prompt makes the instruction explicit. The default frame is 9:16 vertical, which suits TikTok, Reels and Shorts.

Twelve copy-paste text-to-video prompts

Grouped by the four looks. Each prompt is a complete direction; paste it into the studio text field with no photo attached.

Classic bounce

a woman in a plain white t-shirt and denim shorts performs a steady two-step bounce on the beat in a bright bedroom, soft window light from the left, camera locked off at waist height, 12 seconds, 9:16

a man in a grey tracksuit shifts weight side to side in a grounded bounce on a quiet sidewalk at dusk, warm overhead streetlight, camera locked off slightly wide, 10 seconds, 16:9

one dancer in a red hoodie and black pants does a relaxed two-step bounce in an empty basketball court, even fluorescent light, camera locked off at chest height, 8 seconds, 1:1

Hype

a woman in a denim jacket throws one arm up once on the beat in a dark studio, hard rim light from behind, camera pushes in slowly, 8 seconds, 9:16

a man in a black t-shirt hits two sharp shoulder movements on the beat in an empty concrete basement, single hard light from camera left, camera locked off close, 8 seconds, 16:9

one dancer in a yellow windbreaker lands two sharp hits with the shoulders in a graffiti-covered alley at night, one hard white light from camera right, camera pushes in slowly, 10 seconds, 1:1

Smooth groove

a woman in a satin slip dress rolls through a slow groove with weight shifting side to side in a dim living room, warm lamp light from the right, camera drifts left at walking pace, 12 seconds, 9:16

a man in a linen shirt turns slowly in place while stepping through a smooth groove on a rooftop at sunset, golden side light, camera locked off wide, 12 seconds, 16:9

one dancer in a long coat moves through a slow groove in a foggy parking lot, cool blue light from one side, camera locked off at knee height, 10 seconds, 1:1

Cute

a woman in a pastel cardigan does a small bouncy step with both hands at chest height in a white kitchen, bright flat daylight, camera static and slightly high, 10 seconds, 9:16

a man in an oversized beige sweater does a light bouncy step in a sunlit hallway, soft even light, camera static at eye level, 8 seconds, 1:1

one dancer in a mint green sweatshirt taps both feet in a small bouncy rhythm on a wooden floor, warm morning light, camera static low, 10 seconds, 16:9

How to edit a prompt without rewriting it

When a render comes back close to what you want, change one clause and run it again. The five-part structure makes that easier.

1. Identify the clause that caused the problem. If the subject looks wrong, edit the subject clause. If the background is too busy, change the setting.
2. Replace that clause with a new version. Keep the other four clauses identical.
3. Change only one clause per attempt. Two changes at once makes it hard to see which one helped.
4. Test at 480p and 5 or 8 seconds first. That spends 5 or 8 credits instead of 26 or 64.
5. When the motion reads well, re-render the same prompt at 720p, 12 seconds, 9:16 for the final clip.

A table of common edits:

| If you want | Replace this clause | With this clause | | Subject change | one dancer in a grey hoodie | one dancer in a white dress shirt | | Cleaner background | in a dim living room | in a plain white studio | | Warmer light | cool blue light | warm amber side light | | Tighter framing | camera locked off wide | camera locked off close | | Slower motion | hits two sharp shoulder movements | rolls through a slow groove |

Important: the prompt text overrides the look you selected. If you pick Smooth groove but write "fast sharp hits", the text wins and the render follows the sharper action. Choose the look to match the energy you want, or select the look and leave the motion clause out.

FAQ

What does text-to-video cost?

The default render is 720p, 12 seconds, 26 credits. A 480p render costs 12 credits, and a 1080p render costs 64 credits. Longer clips scale from the 12 second table and round up to a whole credit. The studio shows the exact credit total next to the Generate button before you start.

Can I test a prompt without paying?

Each account gets one free preview render at 480p for 5 seconds while the daily preview budget lasts. Google sign-in accounts can also claim 10 free credits per day for three days. Email sign-up starts at 0 credits. There is no unlimited free plan.

Do I need to name the Wipe Me Down song or a specific routine?

No. The model does not reproduce recorded choreography or copy a specific video. Describe the movement in plain words: a two-step bounce, a sharp shoulder hit, a slow groove. That is enough.

How long should a text-to-video prompt be?

One sentence is ideal. Five clauses cover subject, setting, lighting, camera and motion. The text field accepts up to 2,000 characters, but a longer prompt usually makes the motion less readable because the model splits attention across more details.

What is the cheapest way to test many prompts?

Render each idea at 480p for 5 or 8 seconds, which costs 5 or 8 credits. Once one prompt reads well, re-render at 720p for 12 seconds at 26 credits. The pricing page lists the one-time packs and subscription credit amounts. The free preview render is also available once per account while the daily budget lasts.