Crop a 720p 16:9 clip to 1:1: 720x720 clears TikTok's 640x640

A 1280x720 clip cropped to a width of 0.5625 gives 720x720, which clears the 640 x 640 minimum TikTok lists for 1:1 in-feed ads with 80 pixels to spare. Centered, x is 0.21875. Unlike a vertical crop from 720p, the square does not fall below the TikTok minimum.

The numbers

At 720 tall, a 1:1 frame is 720 wide. 720 divided by 1280 is 0.5625. The maximum shift is 0.4375, or 560 pixels. The video filter check is free, so test the request before the paid encode.

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-720p-to-1x1-001" \
  -d '{
    "video_url": "https://media.sume.com/artifacts/artf_demo/talk.mp4",
    "ops": [{ "op": "crop", "x": 0.2188, "y": 0, "width": 0.5625, "height": 1 }]
  }'

Where 720p falls short

The same 720p source cropped to 9:16 would be only 405 pixels wide, below TikTok's 540 x 960 vertical minimum, which is why the square is the safer shape from a 720p master. The video generation docs show Seedance 2.5 offering 720p and 1080p, so generating at 1080p gives more room for any crop.

What Sume does not do

Sume does not upscale a crop to rescue a minimum; a Timeline render at a larger size adds pixels but not detail. It also does not upload to TikTok. Check the spec page again before you submit, since TikTok revises it.

At $0.02 for the crop and $0.10 per output minute for a Timeline render, a 720x720 square is cheap to produce. The only cost to weigh is framing. A 16:9 shot with a centered subject loses little; a wide two-person shot may lose one of them, so view the first frame of the crop before you render.