What finishing a Seedance 2.5 clip costs: trim, detach, compose, join
The edit steps around an AI clip are cheap. Public rates for video trim, audio detach, filter, compose, timeline audio and Timeline render on Sume.

After a Seedance 2.5 generation, the finishing steps on Sume (trim, audio detach, crop, compose, audio join, timeline render) are ffmpeg jobs with no provider inference, and the docs list flat or per-minute public rates for them: $0.02 per trim, $0.01 per detach, $0.02 per filter, $0.02 per compose, $0.01 per audio join and $0.10 per output minute for Timeline. The generation is the cost to watch; the polish around it is cents.
Public rates from the Sume docs
The docs point to GET /v1/catalog for the live rate on each; treat these as the docs' public rates as of 2026-10-05. Video frames and video inspect differ: they bill by their own compute (container seconds at Modal list times 1.25, plus a platform fee), and a transcript adds $0.01 per audio minute.
| Route | Job | Public rate | What it does |
|---|---|---|---|
/v1/video-trim | Trim | $0.02 per job | Cut a [start, end) range into a new MP4 |
/v1/audio-detach | Detach | $0.01 per job | Audio track to wav or mp3 |
/v1/video-filter | Encode | $0.02 per job | dim and crop ops, or a filters-only graph |
/v1/timeline-1.0/compose | Compose | $0.02 flat per job | One still plus one video on screen together |
/v1/timeline-1.0/audio | Concat or split | $0.01 flat per job | Join or slice audio files |
/v1/timeline-1.0/render | Render | $0.10 per output minute, rounded up | Ordered video slots plus an audio spine |
A worked example, in order
Take one 30-second clip and finish it for a vertical placement: trim the first 24 seconds ($0.02), detach the audio ($0.01), swap one line with an audio split and a concat ($0.01 each), crop to a tighter frame ($0.02), and render the final through Timeline ($0.10 for a clip under a minute). The sum of the finishing steps in this example is a few cents; the Seedance job itself is priced by its catalog entry.
- Trim: $0.02
- Detach: $0.01
- Audio split: $0.01
- Audio concat: $0.01
- Crop: $0.02
- Timeline render, under 60 s: $0.10
- Finishing total: $0.17
Budgeting a batch
For a batch, multiply the finishing steps by the number of clips and add them to the generation line. The unbilled plan endpoint for Timeline helps here: run it for each assembly and sum estimated_cost_usd_micros. Remember that the render reserve uses ceil(audio.duration_seconds / 60) minutes, so a 61-second timeline reserves two minutes.
Keep the generation and finishing costs on separate lines in your budget. If finishing costs start to approach a meaningful fraction of the generation, you are probably re-cutting too much, and it is cheaper to fix the prompt and regenerate.
- Sum
estimated_cost_usd_microsfrom plans. - Round up to whole minutes for renders.
- Track generation and finishing costs separately.
What is not in this table
Speech-to-text, captions and audio generation are separate routes with their own rates, so add them if your pipeline uses them. Storage and delivery are outside these job prices. And the generation line depends on the model, resolution and duration, so quote it from the catalog.
Where to read the generation price
Seedance 2.5's per-second price is not in these media-route pages. Read pricing_skus from GET /v1/videos/models for the id, and use the unbilled POST /v1/timeline-1.0/plan for the assembly estimate. Sume bills list price times 1.25 on each video model, per the Video Router docs.
curl https://api.sume.com/v1/videos/models \
-H "Authorization: Bearer $SUME_API_KEY"Sources
Related posts
More in Pricing
- What $5 buys on Gemini Omni Flash 1.1: clips by resolution
$5 buys 41 three-second clips at 360p or 13 ten-second clips at 360p; at 4K it buys 4 three-second clips or 1 ten-second clip. Full table on Sume rates.
- What one localized video costs on Sume: captions, voiceover, dub
Per language, a translated caption job is $0.20, a voiceover on a silent clip up to $0.13, a 60-second dub up to $0.17 and a nine-minute lesson dub up to $1.37.
- Why a 3-second Gemini Omni clip bills 12 cents, not 11, on Sume
Sume rounds each video job up to the cent after the 1.25 multiple: a 3-second 360p Omni Flash clip is $0.12, a 3-second 4K clip $1.13. Row-by-row math.
- How Sume pricing works: plans, one wallet, published model rates
Sume plans set access and concurrency. Usage draws from one prepaid wallet at each model's published USD rate, for generation, the Agent, Formats, and the API.
Written by Sume