How to make a video montage: clips cut to one music track

To make a video montage, choose the music first, decide where each cut falls on it, and then place a short clip between each pair of cuts so the shots change with the track. The music carries the sound, and the clips' own audio is left out. Hard cuts keep the pace, and a short fade or wipe marks a change of section.

By API that is one Sume Timeline render: the track as the sound, the clips as timed slots. The facts below come from the Timeline 1.0 and Music Router docs and Sume's code, read on 2026-09-29.

How do I plan a montage with music?

Plan on the track, not on the clips:

• Pick or generate the track. The Music Router has no length field (duration is rejected); steer the length in the prompt, for example "a 30-second track".
• Listen through and write down the times where you want a cut, such as a drum hit, the start of the chorus, or the last note. A render places each slot at the start you declare, so these times are yours to choose.
• Give each shot the gap between two cut times. Shorter gaps feel faster; a longer shot lets a moment land.
• Choose the part of each clip to show. source_in starts a slot partway into its file, so one clip can supply its best two seconds.

How do I put the clips over the music?

Send POST /v1/timeline-1.0/render with the track as audio.url and its length as audio.duration_seconds, which sets the length of the output. Each video[] slot takes a source_url, an on-screen start, and a duration. The first slot starts at 0 and later starts must increase. POST /v1/timeline-1.0/plan checks the same document for free, without creating a job.

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: montage-001" \
  -d '{
    "audio": { "url": "https://media.sume.com/artifacts/artf_demo/track.mp3", "duration_seconds": 12 },
    "video": [
      { "source_url": "https://media.sume.com/artifacts/artf_demo/a.mp4", "start": 0, "duration": 2.5 },
      { "source_url": "https://media.sume.com/artifacts/artf_demo/b.mp4", "start": 2.5, "duration": 2, "source_in": 4 },
      { "source_url": "https://media.sume.com/artifacts/artf_demo/c.mp4", "start": 4.5, "duration": 3.5 },
      { "source_url": "https://media.sume.com/artifacts/artf_demo/d.mp4", "start": 8, "duration": 4,
        "transition": { "type": "fade", "duration": 0.5 } }
    ],
    "output": { "fade_out_seconds": 1 }
  }'

What happens to the clips' own sound?

It is dropped. In current code a Timeline render takes sound only from its audio track and the optional soundtrack bed, so a montage plays the music alone. If the track is shorter than the montage, declare the length with audio.mode: "silence" and add the music as soundtrack with loop and a fade_out_seconds of up to 10.

What are the limits?

Can I make a montage from my own footage?

Not from files on your computer. Every URL in a render must already be a file in your Sume workspace on media.sume.com, such as a clip Sume generated or an earlier Sume output, and there is no public route for uploading local files. For a montage of generated clips, Can AI make a music video for my song? covers writing a clip for each section, and Image slideshow video API covers stills.

How much does it cost to make a montage?

The render reserves up to $0.10 per output minute, rounded up to the next minute of output, and captures its own compute, never above the reservation, plus a 5.5% agent fee by default. A 12-second montage reserves one minute. The clips and the track are billed by the models that made them; see API pricing.