How to make Bulin 47 AI with Higgsfield, Dreamina or Genematic

The three tools side by side

From each tool's own pages, read on 14 October 2026. Prices and limits change often, so check before you pay.

Higgsfield AI: redraw the real clip with Genjutsu

Genjutsu is Higgsfield's video-to-video tool. Its Motion Transfer mode keeps a clip's movement, camera and timing and rebuilds the people from your reference images, which is why the Higgsfield versions in your feed follow Bulin 47's gestures so closely. Some of them carry #higgsfieldpartner, the tag of Higgsfield's creator programme.

1. 1

   Get the footage. Genjutsu needs the freestyle as a file, between 4 and 30 seconds long. Save it only where the post allows it, and remember the clip belongs to its creator.

2. 2

   Open Genjutsu and pick Motion Transfer. Object Swap is for changing one item in a shot, not a whole person.

3. 3

   Add a full-body photo of yourself. With a face-only photo the model has to invent your clothes, and it often borrows the pink shirt.

4. 4

   Say who replaces whom, for example "replace the man in the pink shirt with the person in the photo", check the credit cost shown and generate.

5. 5

   If the result comes back silent, add the sound in CapCut or your phone editor before you post.

YouTube

Dreamina AI: a new scene in the same style

Dreamina's own Bulin 47 guide suggests its Seedance 2.5 model, your photo, an optional video reference and a prompt that describes the shoulder moves, a partner holding a speaker and a red-lit doorway. It calls the result a recreation, and it says plainly that the page is not a built-in Bulin 47 template and comes with no licensed soundtrack.

1. 1

   Open Dreamina's video generator and choose Seedance 2.5 if your plan includes it.

2. 2

   Upload your photo, or a character you have the rights to.

3. 3

   Describe the scene: one performer, a partner with a portable speaker, a small crowd, a red-lit entrance at night, a handheld vertical camera. Start from Dreamina's sample prompts and edit them.

4. 4

   Generate 5 to 10 seconds, check the hands, the faces in the crowd and the speaker, and run it again with a fix where something looks off.

5. 5

   Add the sound from the app's library when you post.

• What you will not get: Kevin Wai, the real crowd, or Bulin 47's exact timing. Dreamina says a video reference guides the movement but does not guarantee the same gestures.
• What you do get: full control over the look. That suits a cartoon, a costume or a robot better than a straight copy.

YouTube

Genematic AI: a ready-made scenario

Genematic works like a shelf of templates: you pick a scenario, upload one photo and wait about 5 to 8 minutes, according to its site, with no prompt to write. On 14 October 2026 its trending list showed a scenario called "Bulin 47 Freestyle". The listing did not say whether it uses the original footage or the original sound, so look at what it makes on the $1 trial before you pick a monthly plan.

Or skip the editing: one photo into the real clip

The tools above are general video apps, so you do the trimming, the prompting and the sound. Our Bulin 47 video maker only makes this one video: you upload one photo, and you get 15 fixed seconds of Kevin Wai's freestyle with you as the performer, Kevin, the crowd and the camera untouched, and the original sound laid back on. It takes about 4 to 15 minutes and costs 10 credits at 480p (the $9.99 Single pack). Each Google account can first make a free 2-second preview.

The step-by-step version, with the photo that works best and fixes for common problems, is in how to make the Bulin 47 AI video. If you are curious how the different methods compare under the hood, read how the Bulin 47 AI trend works.

Same tools, another trend: Burning Bridges AI

If your feed also shows people dropped into a crowded party scene, that is the Burning Bridges AI trend, built on the music video for Drake's "Burning Bridges" from his 2026 album Iceman (Wikipedia)). Tutorials make it with the same Higgsfield, Dreamina and Kling workflows: a reference clip or a prompt, plus a photo.

The difference is the source. Burning Bridges remixes a music video; the Bulin 47 AI trend remixes a street freestyle filmed on a phone, with a fan, a speaker and a real crowd. Bulin47Video only makes the Bulin 47 one.