Trend explained

What Is the KDA Firestarter Dance? The AI Trend Explained

What the KDA Firestarter dance is, what the "Beat Drop Beat Beat Drop" moment looks like, how people make AI versions from one photo and how to post one responsibly.

7 min read · Last updated: 2026-10-22

Written by the KDA Firestarter Dance team. We rendered this template ourselves, at 9:16 and 720p from a head-to-toe photo of an original character: a render took about four and a half minutes and both attempts kept the face, the hair and the clothes.

The KDA Firestarter dance is the routine that fans associate with Kai’Sa’s solo in K/DA’s FIRESTARTER: voguing hand frames, sharp poses on the "Beat Drop Beat Beat Drop" hits and a dramatic dip. AI video tools now let anyone make a version of it from one ordinary photo, with no choreography to learn.

Key points

  • The dance is known for voguing hand frames, sharp frozen poses and a low, dramatic dip.
  • An AI version starts from one photo: the photo supplies the subject and a template supplies the moves.
  • Label the video as AI generated, and use only photos you have permission to animate.

What is the KDA Firestarter dance?

It is a short-form video trend built around the dance from K/DA's FIRESTARTER music video, and in particular Kai’Sa's solo. Fans talk about its voguing gestures, its sharp poses and a dramatic dip, and about the "Beat Drop Beat Beat Drop" line that the poses land on. Creators use it as a transformation hook: a still photo for the first second, then the dance kicks in.

We cannot point to one origin of the trend, and we have not independently verified view counts or the date it started. What matters for making a version is what the clips have in common, which is the part this tool's template reproduces. K/DA, Kai’Sa and FIRESTARTER belong to their owners, and this site is not affiliated with them.

What happens in the dance?

The template follows a simple arc, described in plain movement rather than copied from any footage.

  1. 1

    The entrance

    Stepping into a spotlight with one hand framing the face

  2. 2

    The voguing frames

    Wrists crossing and arms sweeping, a new pose on every beat

  3. 3

    The beat drop

    Three sharp frozen hits on "Beat Drop Beat Beat Drop"

  4. 4

    The turn and the finish

    A quick turn, a close-up and a final frozen pose

The beats every Firestarter dance video from this template shares.

How do people make an AI version?

There are two routes. In a general AI video tool you write the prompt, describe the movement beat by beat, find the music yourself and often try several times before the face and the body stay consistent. In a template tool, the movement is already described and tested, so you only supply the photo.

Write your own prompt

Flexible, more trial and error

  • You describe every move and when it happens
  • The face can drift between frames
  • You have to add the music yourself

Use the built-in template

One photo, no typing

  • The dance is already described and tested
  • Choose a frame and a quality, then generate
  • A song excerpt comes with the video

Who can star in it?

Yourself, a friend who agreed, or a character you own. A head-to-toe photo works best because the dance uses the whole body. Original character art also works if you made it or have the rights to use it. Photos of minors, celebrities and other public figures are not allowed, and neither are characters or artwork owned by someone else.

Why do voguing poses work well for AI dance videos?

Voguing is built from held shapes: a hand frames the face, a wrist crosses the body, an arm locks into a line, and then the pose changes on the next beat. For an AI video model that is a good fit, because the model has a clear target at every beat instead of a long, blurred transition. In our own renders the sharp frozen poses came out more cleanly than any long, flowing move.

The hard parts are the same as for any AI video: hands, fast turns and close-ups. In our test the face and the clothes survived the turn, but fingers and fast arm movement are where small errors appear first, which is why we recommend watching the whole clip rather than the first frame.

How is the AI version different from the real choreography?

It is a new video in the spirit of the dance, not a copy of it. A fixed silent movement reference and a plain-language prompt describe the moves: a spotlight entrance, a close-up body roll, voguing hand frames, three hits on the beat drop, a fast turn and a fierce finish. The AI then performs them with your subject on a stage it invents, so the timing, the arm shapes and the lighting are never identical to the music video.

That is deliberate. No footage from the music video is copied into your video, we do not rebuild its characters, and you cannot upload your own reference video. The fixed silent reference only guides movement, camera and effects. If you want the real choreography, learn it from the official choreography video; if you want your own face doing something in the same spirit, that is what this tool is for.

Is it safe to post, and what are the rules?

Videos made here are for personal, non-commercial use, and you should label them as AI generated when you share them. The song is copyrighted music that remains subject to the rights of its owners: if you publish a video, you are responsible for following the platform's rules, and a platform may mute, claim or remove a video that contains copyrighted music.

Make your own KDA Firestarter dance video

Upload one photo, choose a frame and a quality and see the price before you generate. Failed renders are refunded automatically.

Make my video →

Frequently asked questions

What is the KDA Firestarter dance?
It is the dance that fans associate with Kai’Sa’s solo in K/DA’s FIRESTARTER: voguing hand frames, sharp poses on the "Beat Drop Beat Beat Drop" hits and a dramatic dip. People now make their own versions with AI from one photo.
Is the "Beat Drop Beat Beat Drop" dance the whole choreography?
No. That line is one moment of the song. The template here is built around that moment and Kai’Sa’s solo, not every part of the full group choreography.
Do I need a prompt to make one?
Not here. The dance is already described and tested. You add one photo and choose a frame and a quality. In a general AI video tool you would write the prompt yourself and may need several tries.
Is this an official K/DA or Riot Games tool?
No. It is independent and not affiliated with or endorsed by Riot Games, K/DA or the artists behind FIRESTARTER. It uses no footage from the music video, and the song remains subject to the rights of its owners.