How to Make the “Three Meters Above the Sky” Trend: AI Club Video
How to make the “Three Meters Above the Sky” trend from two photos: a one-click ready trend, a Seedance 2.5 prompt, whether it can be free and common mistakes.
An AI club video from photos in the style of “Three Meters Above the Sky” takes a single generation: a person on a nightclub balcony looks down, the camera cuts to the dancefloor, and a second person dances in a spotlight in the middle of the crowd. In the IIshki studio you need Seedance 2.5 in “References” mode: upload two photos, call them @Image1 (balcony) and @Image2 (dancefloor) in the prompt, and set 9:16 and 8–12 seconds. Even faster: the ready-made “Gaze From the Balcony” trend already has the prompt written, so you only upload two photos and press Repeat. Below you'll find a copy-paste prompt, five steps, a single-photo option and the mistakes that make faces drift or swap the two characters. If you need to repeat a dance from a specific clip move for move, use Kling Motion Control.
What the trend is and why it works
The story comes from a 2010 Spanish romance film (3 metros sobre el cielo, released in English as Three Steps Above Heaven, which is why the trend also goes by “3 meters above the sky” and “three steps above heaven”): the hero spots a girl from the club balcony, and for a second nothing else exists. In a short video it fits into two shots — the look from above and the dance below. That is why the trend is so easy to recreate with AI: the scene relies on composition, not on the actors.
Three things make the clip recognizable with anyone in it:
- Two shots, not one. First the balcony and the look down, then the dancefloor. The transition is a hard cut or a fast camera fly-down.
- Light singles out the hero. The crowd is dark and blurred, the dancer gets one warm spotlight.
- The look closes the scene. At the end the dancer looks up, and the viewer understands the two have seen each other.
So you can put anyone in the frame: yourself and a friend, a couple, two girlfriends, even a cat on the balcony — the story still reads.
Can you make the “Three Meters Above the Sky” trend for free?
This trend is often searched with the word "free", so to be straight: there is no free video generation here. Choosing the template, the prompt and the settings costs nothing, while the clip itself uses tokens, and the price is on the button before you start. You can spend noticeably less, though, if you don't burn attempts on prompt tuning:
| Route | What you do | When to choose it |
|---|---|---|
| Ready trend “Gaze From the Balcony” | Upload two photos and press Repeat, 480p or 720p | You want a result on the first try |
| Your own prompt in Seedance 2.5 | Copy the prompt below and change the details | You want your own outfit, light or length |
| Draft in Seedance 2.0 | Same prompt, a short 480p clip | Test the idea cheaper, render the final in 2.5 |
Two rules for saving: draft in 480p first and render the final in higher quality; and leave the film title and actor names out of the prompt, because they can cause a refusal and you will have to try again. If a generation fails because of a provider error, the tokens are refunded automatically.
Which model to pick for a club video
| Task | Model | Why |
|---|---|---|
| Classic trend: two people, balcony and dancefloor | Seedance 2.5, “References” | Two photos in one clip via @Image1 and @Image2, up to 30 seconds, 480p–1080p |
| Cheap prompt tests up to 15 seconds | Seedance 2.0, Mini plan | Same @Image approach, up to 9 photos; render the final on 2.5 |
| The exact dance from a specific clip | Kling Motion Control | Transfers motion from a 3–30 s video onto the person in your photo |
| No full-body photo, or you want a ready club frame | Nano Banana Pro | Builds a portrait or a starting frame that Seedance then animates |
If you want to copy not only the composition but also the camera move from a specific scene, see the guide on how to clone a video with AI.
Which photos to upload
Most of the result depends on the photos. Check each one before uploading:
- One person in the frame. In a group shot the model may pick the wrong face.
- The face is visible from the front. No glasses, cap, hand near the face or heavy filters.
- Even light. Hard shadows and colored lighting make facial features harder to read.
- For the dancer — a waist-up or full-body photo. From a face alone the model has to invent the body and outfit, and the dance looks off.
- Your own photos or photos of people who agree. Moderation rejects celebrity photos before generation.
A regular phone photo works: at least 300 pixels on each side, JPG, PNG or WebP up to 30 MB.
Ready-made club video prompt with @Image1 and @Image2
Seedance accepts prompts in English or Russian. Photo references are numbered in upload order: the first photo is @Image1, the second is @Image2.
Vertical 9:16, 15 seconds, photorealistic 35mm film look, light grain, shallow depth of field, handheld with gentle natural micro-movement. Nightclub: a low balcony with horizontal metal railings only one floor (about 3 meters) above a packed dance floor, thin haze in the air. Upper level in cold blue-grey light; the dance floor in deep shadow with one narrow warm white spotlight falling straight down onto @Image2. @Image1 is the person on the balcony, @Image2 is the person in the crowd; both keep exactly the face, hair and outfit from their photos in every shot. 0–6 s: medium shot of @Image1 at the railing, hands resting on the metal, eyes calmly looking down at the dance floor; the camera slowly pushes in at eye level. 6 s: hard cut. 6–11 s: over-the-shoulder shot from behind @Image1, their head and shoulder soft in the left foreground; below, the crowd is clearly visible and @Image2 stands in the center under the spotlight; a slow smooth zoom brings @Image2 closer until the figure fills about a fifth of the frame. 11 s: hard cut. 11–15 s: medium close-up of @Image2 at crowd level, dancing calmly and naturally in the warm spotlight, dark blurred bodies all around; at the very end @Image2 glances up toward the balcony for a moment. Sound: real club atmosphere, bass that grows fuller when the camera reaches the dance floor, crowd murmur, no speech and no music overlay.
What you can change without breaking the scene:
- Light color: blue and red, purple and amber, white strobe.
- Dance: “sways slowly”, “dances with raised arms”, “spins”.
- Ending: instead of the look up — a smile to the camera or a wave.
- Transition: instead of hard cuts, one smooth camera move from the balcony down; keep it under a second or faces start to drift.
- Length: for 10 seconds, tighten the shots to 0–4, 4–7 and 7–10 s.
To swap the roles, don't rewrite the prompt — just upload the photos in reverse order.
More on how @-references and timecoded storyboards work in Seedance 2.5 prompts.
How to make a club video: 5 steps
1. Open Seedance 2.5 and choose “References”
Go to Seedance 2.5 and switch the input mode to “References”. “Frames” mode animates a single starting frame; two different people need references.
2. Upload the photos in the right order
First the person on the balcony, then the dancer. Upload order is what sets @Image1 and @Image2. Mix up the order and the characters swap places.
3. Paste the prompt
Copy the template above and adjust the light, dance or ending. Leave the film title and actor names out.
4. Set format, length and quality
Format — vertical 9:16 for TikTok, Reels and Shorts. Length — 10–15 seconds: shorter and the transition has no room, longer and the dance starts to loop. Run the first test in 480p, and once the scene works, repeat it in 720p or 1080p. The price is shown on the model card before you start.
5. Change one thing at a time
If you don't like the result, change one thing per run: the light, the moment of the transition, the dance description. Change everything at once and you won't know what helped. Drafts are cheaper to debug in Seedance 2.0 on the Mini plan, with the final rendered on 2.5.
The single-photo option
No second photo? The trend still works:
- You on the dancefloor, no face on the balcony. Upload one photo as @Image1 and write: “on the balcony, a silhouette of a person seen from behind, face not visible, looking down”. The rest of the storyboard stays the same.
- You on the balcony, an invented dancer. Describe the dancer in text: age, hair, outfit. The model will invent the face.
- Frame first, then video. In Nano Banana Pro, build a starting frame of “me at the railing of a nightclub balcony” from your photo, then animate it in Seedance in “Frames” mode with the same transition to the dancefloor.
For an exact dance: Kling Motion Control
Seedance repeats the style of the movement, not the choreography step by step. If the dancer should move exactly like in a specific clip, use Kling Motion Control: upload the dance video (MP4 or MOV, 3–30 seconds) and a photo of your character, and the model transfers the motion. With orientation set to “By video” up to 30 seconds are transferred, “By image” up to 10. For the trend it's convenient to do it in two passes: the balcony in Seedance 2.5, the dancefloor in Kling Motion Control, then join the two clips in any video editor. For how Seedance and Kling compare, see Seedance 2.0 vs Kling 3.0.
For an even more precise result, use the Swap mode: upload a 4 to 30 second dance clip and your photo, and the character repeats the poses, expressions and lip movements while the original sound is kept. Kling Motion Control is a working option; Swap gives a better result and is simpler, with no prompt needed.
Common mistakes
| Mistake | Cause | What to do |
|---|---|---|
| Refusal because of the face | Profile, glasses, shadows, filter, a celebrity in the photo | A sharp front-facing portrait in even light, nothing covering the face |
| Content refusal | Film title or actor names in the prompt | Describe the scene in your own words, no titles |
| Characters swapped places | Photos uploaded in a different order than the prompt | Balcony first (@Image1), dancer second (@Image2) |
| The dancer gets lost in the crowd | The crowd is lit as brightly as the hero | Write “the crowd is dark and blurred, one spotlight on the dancer” |
| Faces drift after the transition | No instruction to keep faces, a long camera fly-down | Add “keep both faces in every shot”, keep the fly-down under a second |
| The person on the dance floor is tiny, the face is unreadable | A high balcony and a camera fly-down push the crowd far below | Write “a low balcony, one floor above the dance floor”; instead of the fly-down use an over-the-shoulder shot with a slow zoom, then cut to a medium shot |
| The model picked the wrong person | Several people in the photo | Crop the photo to one person |
| The clip is horizontal | The default 16:9 format was left on | Choose 9:16 before you start |
| The dance loops | Clip too long for one action | 10–15 seconds with three shots, or add a third segment with a new action |
What's next
Start with one 10-second generation in 480p, get recognizable faces, and only then raise the quality. More ready-made scenarios with your photo are in Trends: the prompt is already written, you just upload a picture. A multi-step scene — portrait, starting frame, video — is easy to wire together with nodes in Canvas, and for choosing between video models see the best AI video generators.
Don't want to fiddle with prompts?
Ready-made scenes: upload your photo and get the result in one click.
Models in this article
Frequently asked questions
Which AI makes a nightclub video from photos?
For the “Three Meters Above the Sky” trend the best fit is Seedance 2.5 in “References” mode: upload two photos — the person on the balcony and the dancer — and refer to them in the prompt as @Image1 and @Image2. The model keeps both faces in one clip. Seedance 2.0 works for cheap drafts, and Kling Motion Control for copying an exact dance.
Can I make a club video from a single photo?
Yes. Upload one photo as @Image1 and make that person the dancer, then describe the balcony person in text or show them as a silhouette from behind. Another option is to build a club frame with your face in Nano Banana Pro first and animate it in Seedance in “Frames” mode.
Why does the AI refuse to generate a video with my photo?
Usually the photo itself is the problem: a profile shot, glasses or a cap, a heavy filter, several people in the frame, or a recognizable celebrity. Use a sharp front-facing portrait in even light and remove the film title and actor names from the prompt. If a generation fails because of a provider error, the tokens are refunded automatically.
Can I make the “Three Meters Above the Sky” trend for free?
Not entirely: the template, the prompt and the settings cost nothing, but generating the video itself uses tokens, and the price is shown on the button before you start. The cheapest route is the ready-made “Gaze From the Balcony” trend in 480p: the prompt is already tuned, so you don't waste attempts. If a generation fails because of a provider error, the tokens are refunded automatically.
Should I write “Three Meters Above the Sky” in the prompt?
No, better not. The film title and actor names can trigger a copyright refusal. Describe the scene in your own words — a club balcony, a look down, a cut to the dancefloor, a spotlight on the dancer — and the model will build the same story.
Every model in this roundup is available in the IIshki studio at reduced club prices. Club members get tokens for generations every month.
Join the clubMore articles
- AI Greeting Card and Video Greeting: New Year and Birthday GuideHow to make an AI greeting card and a video greeting for New Year or a birthday: a card with readable text, a talking photo, a festive clip, prompts.
- How to Use Fish Audio: Emotional Text to Speech GuideFish Audio text to speech guide: how to write bracket directions for Drama 3, pick a voice from the library, engine and speed. Ready prompts and fixes.
- ElevenLabs Sound Effects: How to Make SFX with AI (Guide)How to make sound effects with ElevenLabs Sound Effects: a prompt formula, durations up to 22 seconds, seamless loops, prompt adherence and ready prompts.