Caption a 61-second clip: the $0.20 Sume price is for 60 seconds

Sume's video captions page says each accepted standalone caption job captures $0.20, and that this price is for videos of a maximum of 60 seconds under the current fixed estimate. It tells you to read GET /v1/catalog and the OpenAPI spec for the live price. So for a 61-second clip, the docs do not promise $0.20. You have two choices: confirm the live estimate, or trim the clip to 60 seconds first for a flat $0.02, for a total of $0.22.

These facts come from Sume's docs and public catalog, read on 2026-10-09.

What the docs say, line by line

The caption job takes a public HTTPS video_url and returns a captioned video. It is the recommended route when you already have a finished clip. It runs speech-to-text unless you provide script_text, words, cues or segments. A restyle with source_caption_id reuses the earlier word timings, but the price does not change, because a restyle is still a render.

Option 1: trim to 60 seconds first

Video trim cuts [start, end) from one hosted clip into a new MP4 for $0.02. If the extra second is dead air at the end of the clip, this is the cheap, predictable route: $0.02 + $0.20 = $0.22. Check the trimmed clip still ends on a finished sentence, because a cut in the middle of a word produces an odd last caption.

Trim works only on a clip hosted by Sume, so import the video first. Docs state that Sume does not fetch from the open internet for this tool.

Option 2: ask the live estimate

If you cannot cut the clip, check the live price before submitting. The catalog is public and needs no key. Do not assume that a longer clip is billed at the same rate, and do not assume it is billed at a higher one; the docs only say what the estimate was written for.

Whichever you choose, keep the Idempotency-Key the same on a retry so a network error does not create a second paid job.

• 60 seconds or less: $0.20 per the docs.
• Slightly over: trim for $0.02 or verify the live price.
• Much longer: split into parts of 60 seconds or less and caption each.

What not to assume

It is tempting to guess that a 61-second clip costs $0.2033 because 61 / 60 x $0.20 is $0.2033. Nothing in the docs says captions are billed per second. They describe a fixed estimate for videos up to 60 seconds. Treat any other number as unverified until the catalog or a dry check says so.

For campaigns where every clip is 61 to 75 seconds, trimming is not free either: 100 clips would add $2.00 in trim jobs on top of $20.00 in caption jobs. In that case, ask whether the extra seconds are needed at all, since many short-video placements work better under a minute.