How to Make an AI APT MV

This walkthrough covers the whole APT AI video workflow, from picking two photos to downloading the finished clip. There is no prompt to write and no timeline to edit. The scene is prepared, so the only decisions that change the result are the two uploads, the output ratio and the resolution.

Before you start

You need two photos, one per performer, and enough credits for the resolution you want. A single photo of two people will not work, because the singer and the drummer are separate upload slots.

Keep these in mind when you choose the photos:

• One person per photo, facing the camera.
• Even lighting on the face, with no hard shadow across it.
• No sunglasses, hats, or anything covering the face.
• A recent photo, so the identity reads clearly in the render.

Neither person needs to pose like the reference. The singing posture, the drumming motion and the closing dance all come from the template.

Step 1: Upload two photos

In the generator, add one portrait to the singer slot and one to the drummer slot. Each slot takes a single image, up to 8 MB. The uploader marks each file once it finishes, and the counter next to the slot shows how many images are attached.

If you are testing a pairing, a cheap 480P render is the fastest way to check whether the two photos read well together before you spend more credits.

Step 2: Pick the format and resolution

The format selector controls the output canvas:

• 16:9 matches the source clip and suits YouTube and embedded posts.
• 9:16 suits short-form feeds and stories.
• 1:1 suits mixed timelines and square posts.
• Auto follows the template frame when you do not want to choose.

The resolution selector controls quality and cost. 480P uses 50 credits, 720P Standard uses 100, and 1080P Pro uses 200. The required credits are shown in the form before you submit, so there is no surprise at checkout.

Step 3: Submit and review

Once both slots are filled, the generate button submits the paid task. The preview panel switches to a progress state while the provider works, and the finished clip appears in the stage as soon as it is ready. Recent renders stay in the strip underneath, so you can compare two pairings side by side.

When a render is done, download it straight from the stage. The file keeps the template audio attached, so it is ready to post without re-timing.

What to do when a render fails

Provider failures are handled automatically. If a task fails, the unused credits are returned to your balance, and the tile in the history strip shows which task failed. The most common causes are a photo that the moderation layer flags and settings the provider does not support; both are listed in the error message under the preview.

Common questions

Can two people use it as a couple clip? Yes. The two slots are independent, so the template works as a couple apt ai video or an apt duo video ai. You can also upload the same photo twice when you want both performers to share a face.

Does it keep the audio? Yes. The template audio stays attached during generation, so the drum beat and the stage rhythm land on the same beat as the source clip.

Is there a free tier? There is no free generation tier. The cheapest way in is a one-time pack from $9.90, and monthly subscriptions start at $19.

Ready to render one? Open the generator and upload two photos. If you want the background on the trend first, read the APT trend explained.