Sume Timeline cost: the reserve follows audio.duration_seconds

A Sume Timeline 1.0 render costs $0.10 per ceil(output minute). The reserve is ceil(audio.duration_seconds / 60) minutes, so the audio length you declare sets the hold. A 61-second video reserves two minutes. The longest allowed spine is 1800 seconds, which is 30 minutes. The job runs only worker ffmpeg, with no provider inference.

The sums

These are arithmetic from the public rate. The live rate is on GET /v1/catalog, and the docs say to confirm it there.

Ask the plan, do not guess

POST /v1/timeline-1.0/plan runs the schema, the Sume-host URL checks and the pure compiler. It returns object: timeline_plan with duration_seconds, segment_count, billable_minutes, estimated_cost_usd_micros and a filtergraph_summary. It creates no job, reserves no credits and downloads no media. Idempotency-Key is not required.

Send the same body you would send to render. The plan will tell you the cost, and will reject the same invalid documents.

curl -X POST https://api.sume.com/v1/timeline-1.0/plan \
  -H "Authorization: Bearer $SUME_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "audio": { "url": "https://media.sume.com/artifacts/artf_demo/voice.wav", "duration_seconds": 61 },
    "video": [
      { "source_url": "https://media.sume.com/artifacts/artf_demo/intro.mp4", "start": 0, "duration": 61 }
    ]
  }'

Trim the audio, not the clips

Because the rate follows the declared audio length, the cheapest saving is a spine that ends just under a minute boundary. A 61-second spine costs twice a 60-second one. If a cut has one second of slack, remove it from the end of the spine and let the last clip cover it.

What the plan cannot see

A plan cannot predict warnings about short sources that are padded or looped. Those are soft warnings, not failures, and they appear on the real result. If a clip is shorter than its slot, extend the slot or use a longer source.

Coverage can stop at most 0.5 seconds before the end of the spine, so leave a clip long enough for the whole slot.

Other lines on the bill

Material preparation is billed separately. A trim is $0.02, a dim or crop is $0.02, an audio detach is $0.01 and a timeline audio join is $0.01. The captions job is $0.20 for a clip up to 60 seconds.