Tsundere Stomp AI

One photo of an adult does the stomp: they stamp a foot, fold their arms and turn their face away, following the built-in clip move for move.

  • No prompt
  • Sound included
  • No filming
Example · Stomp

PNG / JPG / WebP · up to 10 MB · at least 512px on the short side

0/1 photos

One adult on their own, standing head to shoes with feet and hands in frame and the arms a little away from the body.

Video settings

No prompt needed. Only upload photos of yourself or people who agreed to be featured; celebrity photos and film stills aren't allowed.

What is the Tsundere Stomp, and why is it called the Natsuki stomp?

The Tsundere Stomp is the short clip people film to act fake-offended: one leg lifts and stamps back down, the arms fold over the chest, the face turns away, and the loop repeats until it stops on the crossed pose. It also goes by the Natsuki hmph.

The performer standing with his arms folded across his chest and his face turned aside in a bright room papered with posters and figures

The nickname comes from Natsuki, the prickly-but-soft character in Doki Doki Literature Club: a 5 November TikTok edit of her, captioned “Natsuki cosplay when!?”, is what tied the character to this pose. A tsundere is hard on the outside and soft underneath, and “hmph” is the sound of the sulk the clip copies.

The joke runs the opposite way to a dance trend: nobody is showing off. The person stays straight-faced in an ordinary room, the arms fold like a door closing, and the clip stops dead on the sulk.

The week the Tsundere Stomp trend went around

  1. A TikTok edit of Natsuki, captioned “Natsuki cosplay when!?”, puts the character next to the folded-arms pose.

  2. @savanay_60 posts an adult performing the stomp in a living room; by the 12 November snapshot it has 357.8K likes and 154.6K shares.

  3. @matheogoose posts a second adult take with the same caption; 300.3K likes and 117.5K shares in the same snapshot.

  4. @user8043782126795 posts a CapCut production tutorial for the same moves, the point where people stop needing to film themselves.

Sources: TikTok posts, their captions and the sound page, and the Doki Doki Literature Club character page; dates, likes and shares as read on 12 November 2026.

One loop of the stomp, in four moments

One adult, one unbroken front-on take of 7.41 seconds. Here is what the performer copies, in order; the caption people post it with is “tsundere stomps her foot”.

  1. The performer with one leg lifted forward, the sole of his shoe toward the camera10:00–0:01

    The leg comes up, the foot stamps down

    A leg lifts forward with the sole toward the lens and comes back down. A photo without floor below the feet leaves this landing for the model to invent.

  2. Arms folded across the chest with the head turned down and to one side20:01–0:02

    The arms fold, the face turns away

    The arms swing up and the forearms cross over the chest, and on the same beat the chin drops and the face turns aside: the sulk the trend is named after.

  3. Standing with both arms held out to the sides and the feet apart30:02–0:05

    A second stamp, then the arms fly open

    The loop runs again: another lift and stamp, then both arms swing out to the sides before the next fold. This is the widest the body gets on screen.

  4. Arms folded again at the end of the take, standing still40:05–0:07

    The arms cross and the pose holds

    The take ends on the folded arms with the face still turned, held still for the last second of the clip.

How to make a Tsundere Stomp AI video from one photo

Nothing to film and nothing to write: the stomp, the folded arms and the sound are built in. The one photo decides who performs it, in what clothes and where.

  1. The performer standing full body in a black outfit and white gloves, arms at his sides

    Photo 1Dancer

    The adult in your photo takes the part: their face, hair, clothes and room are kept, and the stomp, the arm cross and the turned face are copied from the built-in clip.

    • Standing, head to shoes, with the feet in the frame
    • Arms held a little away from the body

Photo checklist

Works well

  • Woman in a denim jacket and jeans standing head to shoesWhole body, head to shoes
  • Woman in a navy sweater holding both open hands out in front of herHands clear of the body, so the fold reads
  • Man in a grey sweater and dark jeans standing head to shoesOne adult in the photo, nobody else

Avoid

  • Close selfie of a woman on a street, face onlyA chest-up selfie with no feet in it
  • Two men side by side in one photo at a cafeTwo people in one photo
  • Man seated on a sofa holding a mug, body hidden from the waist downSitting or lying down

From a photo to a post

  1. Pick one full-body photo

    Drop one picture into the dancer slot, taken far enough back that the whole person and the floor under them are in frame.

  2. Click Generate

    The button shows the exact price before any credits are spent, and 720p or 1080p is set just above it.

  3. Post it

    Download the MP4 and post it as it is: the clip’s sound is already on it, and the frame is already vertical.

Pets do not work here: a dog or cat has no arms to fold and no way to stamp one foot, so this template takes one adult per photo.

Tsundere Stomp, or a tsundere voice or chat tool?

Searches around “tsundere” pull in voices, translators and chat apps. Only the first row below is a video you can make here.

TermWhat it isHere
Tsundere Stomp / Natsuki hmphOne photo performs the stomp and the arm crossThis page, from your photo
Natsuki StompThe character edit that named the poseA nickname, not a costume
tsundere AI voiceA generated prickly voice lineNot offered here
tsundere translatorRewrites a sentence in a tsundere toneNot offered here
tsundere AI chatRoleplay chat with a tsundere characterNot offered here

Any outfit, any room: Tsundere Stomp videos from your photo

Four people, four rooms, four outfits, and the same four moments in each clip. Read them as casting options, not as versions of the effect.

  • Young woman in a cream hoodie and denim shorts standing in a bright living room
    Hoodie

    Sunday hoodie, living room

    Caption: nobody talk to me today

    A loose hoodie is fine as long as both hands are out of the sleeves.

  • Man in a light blue shirt and dark trousers standing in a bright home hallway
    Work shirt

    Work shirt, still in the hallway

    Caption: clocked out, still annoyed

    Shirt and trousers already match the straight-on stance; a hallway gives the arms room.

  • Woman in a navy tracksuit standing in a bedroom in front of a white wall
    Tracksuit

    Tracksuit, bedroom wall

    Caption: the mood survived the workout

    A tracksuit keeps arms and legs readable, which is all the arm cross needs.

  • Older woman in a long camel coat standing in a bright apartment entryway
    Coat

    The long coat, by the door

    Caption: dressed for the occasion, done with it

    Age changes nothing in the move; a long coat only needs the arms clear of the pockets.

Tsundere Stomp AI prompt: run the stomp in a reference tool

This page takes no prompt: the stomp is built in. Where the tool takes a video reference, attach the clip as @Video1 and your photo as @Image1, and the text decides which file gives the moves and which gives the person.

Reference clip

Prompt length 906

Use @Video1 as the motion reference and @Image1 as the performer. The adult in @Image1 stamps and folds their arms in the place @Image1 was taken in: keep their face, hair, clothes, body and that room with its light.

Copy the moves in @Video1's order and on its beats: one leg lifts forward and the foot stamps back down; the arms swing up and fold over the chest as the face turns away; a second lift and stamp, then both arms open wide; the arms fold again and the crossed pose holds until the clip ends.

Take only the moves, the timing, the camera and the sound from @Video1, never its dancer, his black-and-white dress, his white gloves or the poster wall behind him: one fixed front camera at chest height, one unbroken 9:16 take, and keep @Video1's original sound. Both arms stay separate from the chest, five fingers per hand, both feet on the floor, one person only, no animals, no text or logos.

Worth spelling out

  • The moves in the clip’s order

    Stomp, fold and turn away, second stamp, arms open, then the hold: the arm cross only lands on the beat if the order is there.

  • Fix the camera in words

    “Fixed front camera, one unbroken take, 9:16” stops a tool from adding the cut or the push-in this clip never has.

  • Point at the sound

    Ask for @Video1’s original sound by name; the text cannot rebuild the track the stomps land on.

Leave out

  • A request for a zoom or a cut

    Any camera move you add pulls the take away from the front-on loop, and the folded arms drift off the chest.

  • The template’s dress or gloves

    Naming the dress, the apron or the white gloves puts that costume on your photo instead of your own clothes.

  • Silence about the room

    Left unsaid, the poster wall of @Video1 can follow the moves into your photo.

A one-line ask and a usable one

Too vague

Make a tsundere stomp video with my photo.

Works

Use @Video1 as the motion reference and @Image1 as the performer: they stamp, fold their arms and turn their face away in their own room, with @Video1's fixed camera, timing and original sound.

The working line names both files, the four moves in order and what stays: the room, the camera and the sound.

Tsundere Stomp AI generator: length, resolution and credits

Tsundere Stomp AI generator: length, resolution and credits
Photos
1 photo, one adult
Photo files
PNG / JPG / WebP, up to 10 MB, at least 512 px on the short side
Length
7.41 seconds, the length of the built-in clip
Frame
9:16, vertical
Resolution
720p or 1080p; 720p by default
Sound
Keeps the built-in clip’s sound
Model
Kling 3.0 Motion Control
Credits
193 credits at 720p, 252 at 1080p
Visibility
Public on a free account; paid plans can keep it private
If a run fails
The credits go back to your balance automatically

When the stomp comes out wrong

Folded arms, a stamp toward the lens and a face that turns away are the three hardest moments here. These are the shapes the misses take, and the change that helps most.

  • The hands fuse into one another

    Why it happens
    Two forearms overlap on the chest and both gloves are white, so fingers merge where the arms cross.
    What to change
    Keep the hands clear of each other in the photo and check the cross frame, not the whole clip; bare arms or sleeves read cleaner than long gloves.

  • The foot drifts and the leg stretches

    Why it happens
    The stamp comes toward the lens, so the lifted leg is the closest thing on screen and grows as it approaches.
    What to change
    Use a photo taken further back, with floor space below the feet, and compare the foot at the lift with the foot at the landing.

  • The face changes after the turn

    Why it happens
    The head turns away while the arms already cover the chest, which is the moment the model has least face to hold on to.
    What to change
    Shoot from face-on, with the hair off the face, then compare the first and last frames rather than the fast middle.

  • The template’s dress or room arrives, or a second body

    Why it happens
    The take brings an outfit and a room walled with posters, and a pet or a second person in your photo gives the model one more body to keep.
    What to change
    One adult on their own against a simple background, with the outfit you want in full view: that is what the result copies. If you run the prompt route instead, describe those clothes in the reference-tool prompt.

More trending AI video effects

Every template works the same way: add your photos, press Generate and get a shareable video without writing a prompt.

View all effects

FAQ

Tsundere Stomp AI questions

Is the Tsundere Stomp AI free?+

No. Every run costs credits; the amount depends on the resolution you choose, and the Generate button shows the exact price before you spend anything. New accounts get 10 credits at sign-up.

Do I need a prompt for the stomp?+

No. There is no prompt box on this page: the moves, the camera and the sound are built into the template, so there is nothing to copy from another site either.

Can my cat or dog do the Tsundere Stomp?+

No. The move is a standing stamp with folded arms, and this page is built for one adult per photo; a pet gives the model a body that cannot fold its arms.

Does the video keep the sound?+

Yes. The result carries the built-in clip’s audio, so the stomps land on it. That audio is the clip’s own sound; no track title is confirmed for it, so nothing beyond the clip itself is promised.

How long is the video, and can I make it square or longer?+

It runs 7.41 seconds and keeps the clip’s vertical 9:16 frame. Length and frame follow the clip, so there is no 15-second or square version; resolution is the only setting, 720p by default and 1080p as an upgrade.

Is this an official Natsuki or Doki Doki Literature Club effect?+

No. Natsuki is a character in Doki Doki Literature Club, and this template copies a real person’s moves, not her look. Oeneo AI is an independent tool; the game and its makers are not involved.

Why does the face look different after the head turn?+

The turn happens while the arms already cover the chest, the hardest moment in the clip for a face. A photo taken from the front, with the hair off the face, helps; compare the first and the last frame rather than the fast middle.

How is this different from Motion Control?+

Motion Control in the video tools takes any clip you upload and lets you set it up. This page fixes one clip, so there is nothing to upload and nothing to choose beyond resolution.

Whose photo can I use?+

Your own, or a photo of another adult who has agreed to it.

Tsundere Stomp

Make your Tsundere Stomp video

One full-body photo in, a 7-second stomp with folded arms and the clip’s sound out.

Make the stomp videoSee the four moments