Crop 21:9 to a 1:1 square: width 0.4286, centered x 0.2857

A 2520x1080 ultrawide clip becomes 1080x1080 with a crop of width 0.4286. TikTok's in-feed page lists 1:1 at a 640x640 minimum. Request and numbers inside.

4 min readSume
All posts

To turn a 21:9 clip such as 2520x1080 into a 1:1 square, keep the full height and crop the width to 0.428571, which is 1080 of 2520 pixels. Centered, x is 0.285714. 1080x1080 is well over the 640 x 640 minimum TikTok lists for 1:1 in-feed video.

The numbers

A 1:1 frame at 1080 tall is 1080 wide. 1080 divided by 2520 is 0.428571, and the maximum shift is 0.571429, or 1440 pixels. The video generation docs list 21:9 among the ratios, and each model shows its own subset, so check supported_aspect_ratios first.

Crop windows for a 2520x1080 source to 1:1, our arithmetic (read 2026-10-07)
WindowxPixels kept (left to right)width
Left-aligned00 to 10800.428571
Centered0.285714720 to 18000.428571
Right-aligned0.5714291440 to 25200.428571

The request

Crop fractions are applied to the source frame, so the pixel result depends on the source size. The video filter docs require x and y in [0, 1], width and height in [0.05, 1], x + width at most 1 and y + height at most 1, and the compiler rounds to even values. One crop is one $0.02 encode job, the check is free, the source must be 300 seconds or shorter, and the request needs an Idempotency-Key. Sume does not auto-detect faces or subjects, so you choose the offset.

curl -X POST https://api.sume.com/v1/video-filter \
  -H "Authorization: Bearer $SUME_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: crop-21x9-to-1x1-001" \
  -d '{
    "video_url": "https://media.sume.com/artifacts/artf_demo/talk.mp4",
    "ops": [{ "op": "crop", "x": 0.2857, "y": 0, "width": 0.4286, "height": 1 }]
  }'

What you give up

A square keeps 43 percent of the ultrawide width. A wide establishing shot loses most of its context. For a product shot with the subject centered the loss is small, and you can move x to follow the subject instead.

What Sume does not do

Sume does not track the subject, so a moving subject needs a fixed window or several trims. It does not upload to TikTok Ads either. TikTok also lists 9:16 as the recommended vertical ratio, so check whether a square is the right shape for the placement.

A good test is to look at a frame from the start and the end of the crop before paying for the render. If the subject drifts out of the window, split the clip with video trim at the point of movement and crop each half with its own x. Each trim is $0.02 and each crop is $0.02, so a two-window square costs $0.08 in Sume jobs: two trims and two crops.

Sources

Related posts

More in Media tools

All Media tools posts

Written by Sume