Sume video captions cost a flat $0.20 per clip up to 60 seconds

A standalone Sume caption job costs $0.20, whatever the clip's length up to 60 seconds. The captions docs say each accepted job reserves and captures that amount, and that the price is for videos of a maximum of 60 seconds under the current fixed estimate.

Batch math

Because the price is flat per job, a 10-second clip and a 55-second clip cost the same. Captioning is cheapest per second when the clip is near the limit.

Where captions are not separate

The avatar-video docs say inline captions do not create a separate billed video-caption job; they are part of the avatar estimate instead. So if you generate a talking avatar and ask for captions in the same request, you do not also pay the $0.20 caption job. Check the avatar docs for the add-on rate.

Flat pricing changes the order of work

Since cost does not depend on length below the limit, the cheapest plan is to caption the final cut, not the pieces. Three 20-second clips captioned separately cost $0.60; the same footage cut into one 60-second video and captioned once costs $0.20. If your timeline render comes first at $0.10 per started minute, the combined cost of render plus captions for that minute is $0.30.

The docs recommend reading GET /v1/catalog for the live price, and note that the $0.20 is the current fixed estimate. If you caption at scale, read the catalog at the start of each run.

Cost decisions

• Burn captions in once, at the end. Captioning a draft you will re-edit pays $0.20 twice.
• Restyling an earlier caption is a new job. The request accepts a source_caption_id, but each accepted job is $0.20.
• Clips over 60 seconds fall outside the stated flat price. Look up the live rate in GET /v1/catalog before you caption long videos.