The clip is everywhere right now: someone leaps off a high cliff, free-falls for a second, lands on a trampoline and gets launched back up to the exact spot they jumped from. It looks filmed, but almost every version you see is AI-generated from a single photo. You can make one yourself: upload a photo to a fixed trampoline template, press generate, and get a finished 15-second vertical MP4 with the crowd and wind sounds already in it.
That is a real output — one photo in, that whole clip out.
What is the AI trampoline video trend?
It started as one real video: a daring jump off a cliff onto a trampoline hanging over the water, ending with the jumper bounced back to the top. The version flooding TikTok, Reels and Shorts now is AI: a model re-renders that same fixed clip with a new person in it — your face, hair and outfit come from the photo you upload, while the cliff, the camera move and the impossible bounce-back stay identical.
There is a second flavor of the same trend: animals on trampolines — bunnies, rabbits, bears, deer. Same idea, different jumper. This is why every AI trampoline video you scroll past looks shot-for-shot the same: creators are not refilming anything, they are all feeding photos into the same fixed template.
How do you make one from a photo?
Four steps, no video skills needed:
- Pick a clear, front-facing photo. One person, even lighting, no sunglasses or masks — the model builds the jumper out of this photo, so a covered face cannot transfer. Full-body vertical photos fill the frame best.
- Upload it on the Trampoline Jump AI page. The photo goes through a content check and is expanded to a full-body reference if needed, so the jumper looks whole in the clip.
- Press generate and wait. The template carries the cliff, the camera and the bounce physics — you are not prompting anything. The crowd reacting to the jump and the wind rush of the fall are baked into the audio automatically.
- Download the MP4 and post it. You get a 720P, 9:16 vertical file, ready for TikTok, Reels or Shorts as-is — keep the original sound or overlay a licensed sound from the platform's library on top.
Generation runs on the AI trampoline video generator page, and the result also lands in your account's My videos section, so nothing is lost if you close the tab.
What about the bunny and bear versions?
The animal clips — bunnies bouncing, a bear dropping off a diving board, a deer mid-air — come from the same workflow applied to animal subjects: a fixed trampoline clip plus a reference image of the animal. Some generic AI video tools can attempt them from a pet photo with a text prompt, but results are far less consistent than the trend's human template, and most one-click trampoline templates are built around people.
One honest note: this site's trampoline tool makes the human cliff version only. It is built around photos of people and does not accept pet photos — if you want your dog on a trampoline, that is a different (and much less predictable) generation job than the trend template everyone is copying.
Which tools can make it?
The trend is popular enough that several tools offer it as a one-click template. A factual comparison:
| Tool | What you get | Pricing model |
|---|---|---|
| AI Video Creator (this page's tool) | The cliff trampoline clip, 15 seconds, 720P vertical, crowd and wind sound baked in | $7.9 per video, or 1 credit from a pack |
| CapCut Dreamina | "High-altitude trampoline" effect inside the CapCut/Dreamina ecosystem | Credit-based — check their site for current pricing |
| MakeViral | "High Altitude Trampoline" one-tap template | Credit packs — check their site for current pricing |
| LightX | Photo-to-trampoline workflow in their editor | Subscription — check their site for current pricing |
Whatever tool you pick, the recipe is the same: one clear photo of a person, one fixed template. Differences show up in output length, sound and what happens when a generation fails.
What does it cost?
The trampoline template is $7.9 for one video, or 1 credit from any regular pack ($9.9 for 2 videos, $19.9 for 5, $39.9 for 11 — packs never expire). There is no free preview for this template; the one-time free preview (768p, 5 seconds) belongs to the text-to-video tool on the home page. If a trampoline generation fails, the credit is returned to your account automatically.
How long does it take?
Usually 5–15 minutes from pressing generate to a finished clip. You do not need to keep the page open — the video waits in My videos with playback and download, and you are not charged again when you come back to check.
One honest limitation
The template is one scene: the cliff, the drop, the bounce back. That is the point — it is why every result nails the physics the trend is famous for — but if you wanted a different location or your own footage, this tool will not produce that. Photos of public figures are blocked, uploads must be photos you have the right to use, and heavy filters or covered faces weaken the likeness. Output is fixed at 15 seconds, 720P, 9:16 vertical.
The takeaway
The AI trampoline video is a fixed template doing the heavy lifting — the cliff, the camera and the impossible bounce-back are the same in every version, and only the jumper changes. That is exactly what Trampoline Jump AI automates: one photo in, the trend's clip out, with sound, for $7.9 or 1 credit.
For the broader workflow around short-form AI clips, start with how to make a TikTok video with AI, or see what an AI video creator is for the full picture of what these tools do and do not do. Weighing tools by billing model? AI video generators without a subscription compares packs versus plans in plain numbers.
By AI Video Creator Team