Vertical and square cuts of one 45-second launch video: $0.60

To get a vertical and a square version of one 45-second product launch video, render the same voiceover and clips twice with Sume Timeline 1.0: once at 1080x1920 and once at 1080x1080. Each render is $0.10 because the public rate is $0.10 per ceil(output minute) and 45 seconds rounds up to one minute. Add one caption job per cut at $0.20 and the pair costs $0.60 (read 2026-10-08).

Why two cuts, and which placements accept them

YouTube says a Short can be up to three minutes and can use a square or vertical aspect ratio. TikTok recommends 9:16 for non-Spark ads, with a minimum of 540x960, and accepts 1:1 at a minimum of 640x640. A 1080x1920 file and a 1080x1080 file both clear those minimums, so a small team can post the same launch story to both feeds without reshooting.

What the two renders cost

The arithmetic below uses the Sume docs rates for Timeline 1.0 and standalone captions. Captions are a fixed $0.20 for videos of up to 60 seconds, which covers a 45-second cut. The Timeline job uses no provider inference, only worker ffmpeg, and the plan call is unbilled.

The two request bodies

Both renders reuse the same spine and slots. Only the output block changes. Set fit per slot: cover fills the frame, contain keeps the whole picture, and blur fills the sides with a blurred copy. Import every file first, because Sume rejects off-host URLs. Send Idempotency-Key on every render, and a different key per cut.

Run POST /v1/timeline-1.0/plan with each body before you pay. It returns billable_minutes and estimated_cost_usd_micros, so you see the $0.10 before you render.

curl -X POST https://api.sume.com/v1/timeline-1.0/render \
  -H "Authorization: Bearer $SUME_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: launch-square-001" \
  -d '{
    "audio": { "url": "https://media.sume.com/artifacts/artf_demo/vo.wav", "duration_seconds": 45 },
    "video": [
      { "source_url": "https://media.sume.com/artifacts/artf_demo/hero.mp4", "start": 0, "duration": 20, "fit": "cover" },
      { "source_url": "https://media.sume.com/artifacts/artf_demo/demo.mp4", "start": 20, "duration": 25, "fit": "blur",
        "transition": { "type": "fade", "duration": 0.25 } }
    ],
    "output": { "width": 1080, "height": 1080 }
  }'

Caption each cut, not the source

Standalone captions take a public HTTPS video_url, so run them on each rendered MP4. If the launch voiceover is English, slam is the default for Latin text. For Korean speech use a Hangul style such as korean-ad, because Latin styles return caption_hangul_text_latin_style instead of burning tofu boxes.

Keep the caption job on the finished cut so the words sit inside the safe area of that aspect ratio.