Build a 3-minute vertical Short from 30 s AI clips with Timeline

YouTube Help lists Shorts up to three minutes, square or vertical (read 2026-10-08). On Sume that is six 30-second clips, since wan-3.0 and seedance-2.5 both reach 30 seconds per request (Video generation), joined by one Timeline 1.0 render of 180 seconds.

Cost of the six clips plus the join

Clip prices from the pricing estimators at 9:16; the render is 3 billable minutes.

Render request

Six slots at starts 0, 30, ... 150 over a 180-second spine. Only the first two are shown to keep it short; extend the list.

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: short-180-001" \
  -d '{
    "audio": {"url": "https://media.sume.com/artifacts/artf_demo/vo.wav", "duration_seconds": 180},
    "video": [
      {"source_url": "https://media.sume.com/artifacts/artf_demo/s1.mp4", "start": 0, "duration": 30},
      {"source_url": "https://media.sume.com/artifacts/artf_demo/s2.mp4", "start": 30, "duration": 30,
       "transition": {"type": "fade", "duration": 0.5}}
    ]
  }'

Notes

• Declared starts are authoritative, and coverage can fall short of the spine by at most 0.5 s, so the last slot must reach 180 s.
• Six slots is under the 12-slot point where the auto render strategy starts chunking.
• Run /v1/timeline-1.0/plan first; it is free.

Limits that matter at 180 seconds

Timeline 1.0 accepts an audio.duration_seconds of 1 to 1800 and 1 to 200 video[] slots, so a three-minute Short is far from the ceiling. The default output is 1080 by 1920 MP4, which is the vertical case; a square Short needs output.width and output.height.

If you want music under the voice, add a soundtrack with duck_db so the bed drops while the narration speaks. Ducking needs a real audio spine, not silence (Timeline 1.0).