Tutorial · updated October 15, 2026
How to make the Bulin 47 AI video, step by step
This tutorial shows how to make the Bulin 47 AI video that stays closest to the original clip: the same 10-second scene, the same cuts and camera moves and the same sound, with your face in it. You need 1 photo and no prompt, editing app or download of the TikTok.
Switch between the original clip and a real result of this maker, unedited, made October 11, 2026 with photos of AI-generated sample people.
What you need
- 1 photo: The freestyler: stands in the centre in the black T-shirt with a chain.
- 100 credits ($9.90 in the smallest pack), taken only when the video finishes.
- about 5–10 minutes of waiting. You can leave the page.
- A phone or a computer with a browser. No app, no editor, no account on another AI tool.
Step-by-step: make the Bulin 47 AI video
Open the template
Go to the Bulin 47 AI trend page on Remake Your Meme. The maker sits next to a real 10-second result, so you can compare it with the original clip before you spend anything.
Choose 1 photo
The Bulin 47 AI video needs 1 photo: photo 1 is the freestyler (stands in the centre in the black T-shirt with a chain). Use one person per photo with the whole face visible, ideally front-facing and sharp.
Upload and confirm
Drop the photos into the maker and confirm that you have permission to use them. They go straight to private storage and are deleted with the video after 30 days.
Keep the trend sound
Leave "Trend sound" on to get the audio of the original clip under your video. Choose "No song" if you want to add the official sound inside TikTok instead.
Start and wait
Press "Make my video". The Bulin 47 AI video usually takes about 5–10 minutes. You can close the tab or start another one; the video waits in Your videos, and we email you when it is ready.
Download and post
Download the 10-second 9:16 MP4 without a watermark and post it on TikTok, Reels or Shorts. Label it as AI-generated where the app asks, and tag the people in it.
Why this remake stays almost identical to the original
Most AI versions of this trend are generated from a prompt and a reference image, so the room, the timing and the camera drift away from the clip. Here the original clip itself is given to the video model as the motion reference, together with your photos, so every shot follows the original and only the people change.
Play the original and the remake together
Both clips start from the same second, so you can follow every shot and cut side by side.
What stays exactly as in the original
- The red-lit entrance, the man holding the speaker and the crowd filming on their phones.
- The freestyle moves, the mouth movement, the cuts and the camera of the original 10-second clip.
- The freestyle audio, in 9:16 for TikTok.
What changes
- The freestyler's face and hair come from your photo, in every shot.
- The outfit stays as in the clip.
How to get the closest result
- Use one clear, front-facing photo of one person; short or tied-back hair reads best under the red light.
- This template runs on a stronger model than the 16:9 ones, because a lighter model kept the original freestyler in some shots in our tests.
Three ways to make a Bulin 47 AI video almost identical to the original
| Method | How close to the original | What you do |
|---|---|---|
| This maker | Very close: built from the original clip, with the scene, cuts, camera and sound already set up. | Add 1 photo and press one button. Failed videos are refunded. |
| Motion transfer, such as Higgsfield Genjutsu | Very close: Higgsfield says the motion, camera and timing stay as filmed and everything else is rebuilt from your references. | Find and download the original clip, upload it with your photos, describe the swap, check every shot, then add the sound back. |
| Prompt and a reference photo | Loose: the model invents the room, the timing and the camera again on every run. | Write a long prompt for every shot and retry until the scene looks right. |
Higgsfield's description read on its Genjutsu page on 15 October 2026. We are not affiliated with Higgsfield.
The manual way, for comparison
This is how people make the Bulin 47 AI video with a general AI video tool. It works, but every step is a place where the result drifts from the original:
- Save the freestyle clip from TikTok and trim it to the ten seconds you want.
- Give the clip and a photo of your face to a video-to-video AI model that takes a reference image.
- Check every shot: in our own tests on 10 October 2026 the first model kept the original freestyler in some shots, so expect to run it more than once.
- Put the freestyle audio back under the video in an editor such as CapCut and export it at 9:16.
Photo tips for the Bulin 47 AI video
- One photo, one person: a clear, front-facing face. The crowd around you stays as it is.
- Short or tied-back hair reads best under the red light; the outfit comes from the clip.
If you are looking for this video by description
People often ask for the Bulin 47 AI video without its name. These all mean the same clip, and this tutorial covers them:
- “the AI video of a guy freestyling to a speaker under red lights”
- “Bulin 47 freestyle AI with my face”
- “dembow freestyle crowd filming AI trend”
Read more about where the trend came from on the Bulin 47 AI trend page, or browse all meme video templates.