Soundtrack for a 3-minute Short: one prompted track or a looped bed
Music Router is $0.125 per generation whatever the length, so a 3-minute Short pays the same for one prompted track or a 30 s bed looped by Timeline soundtrack.

A 180-second YouTube Short can take its music either from one Music Router generation or from a short bed that Timeline loops. Both cost $0.125 for the track, because Sume charges a flat $0.125 per music generation on every music model (Lyria 3.5 by default), plus $0.30 for the three Timeline render minutes: $0.425 either way.
Shorts run up to 3 minutes (YouTube Help, read 2026-10-09). The choice between the two options is a quality call, not a price call.
Two ways, same bill
Music Router rejects duration and duration_seconds, so the length goes in the prompt (for example, a 3-minute track with section timings). Timeline's soundtrack object has a loop flag and a fade_out_seconds of up to 10.
| Option | Music generations | Music cost | Timeline (3 min) | Total |
|---|---|---|---|---|
| One track prompted to 3 minutes | 1 | $0.125 | $0.30 | $0.425 |
| 30 s bed, soundtrack.loop true | 1 | $0.125 | $0.30 | $0.425 |
| Three 1-minute tracks as audio.parts | 3 | $0.375 | $0.30 | $0.675 |
Request the track, then place it
Generate the music, import the finished file to media.sume.com, and pass it as the audio spine (a single track as spine needs no soundtrack object). If narration is the spine, use the music as the bed and set duck_db to lower it under the voice; ducking needs a real spine, not silence.
curl -X POST https://api.sume.com/v1/music-router/generate \
-H "Authorization: Bearer $SUME_API_KEY" -H "Content-Type: application/json" \
-H "Idempotency-Key: short-bed-001" \
-d '{"model":"sume/music-auto","prompt":"Warm lo-fi, 84 BPM, instrumental. A 3-minute track, calm intro 0:00-0:20."}'
# in the Timeline body, as a bed under a voice spine:
# "soundtrack":{"url":"https://media.sume.com/artifacts/artf_demo/bed.mp3","gain_db":-12,"loop":true,"fade_out_seconds":4,"duck_db":10}Gotchas
- A prompted length is a request, not a guarantee. Check the finished duration before you plan around it.
- soundtrack.fade_out_seconds is at most 10, and edge fades together cannot exceed the output length.
- Looping a 30-second bed six times is audible; if the Short has quiet sections, prompt for a 3-minute track instead.
- If the music should be the only sound, send generate_audio false when you generate the clips.
Sources
Related posts
More in Media tools
- Source sizes that reach 3840x2160 within Sume upscale's 4x limit
Sume video upscale accepts factors 1.1 to 4. 960x540 needs 4, 1280x720 needs 3, 1920x1080 needs 2; 854x480 tops out at 3416x1920. Same $0.009 a second.
- Cut a 180-second Short into three 60-second trims for $0.06 on Sume
YouTube blocks Shorts over one minute that carry an active copyright claim. Three Sume video trims of 60 s each cost 3 x $0.02 = $0.06 and stay at one minute.
- Three caption styles on one clip for $0.60 with source_caption_id
Standalone captions are $0.20 per job for clips up to 60 s. Restyling with source_caption_id reuses the word timings, so three looks cost 3 x $0.20.
- TikTok 516 kbps floor: 0.033 bits per pixel at 540x960, 0.008 at 1080p
At 30 fps TikTok's 516 kbps floor is 0.033 bits per pixel at 540x960 but 0.008 at 1080x1920. Why a bigger frame needs a higher real bitrate to look the same.
Written by Sume