JoggAI API 15/30/60 s lengths vs Sume's 4 to 60 s avatar window

JoggAI's API documentation text lists 15, 30 and 60 seconds as script-length options in its AI script generation section, plus a custom-audio maximum of 10 minutes. Sume's Avatar 1.0 talking-video route takes any script or scene plan that Sume estimates at 4 to 60 seconds inclusive, so a 22-second script is a 22-second request with no length bucket in between.

The comparison below comes from the JoggAI API docs and JoggAI pricing page, both read on 2026-10-03, and from Sume's Generate avatar video page and API pricing. Where a page was silent, this post says so instead of guessing.

What do the two sets of docs say about length?

The JoggAI docs text lists scripts timed for 15, 30 and 60 second videos (its example sends video_length_seconds of 15) and says custom audio can run to 10 minutes. It also advises avoiding very long scripts for photo avatars, under two minutes. The 15/30/60 list sits under script generation, so read it as a menu of script lengths, not as a stated ceiling for every avatar video.

The JoggAI pricing page speaks about a different thing: the longest video a plan can produce in the app. It lists 3 minutes on Free, 30 minutes on Starter and Creator 1, and 60 minutes on Team. Those are plan ceilings, not API clip lengths, and they should not be read against the 15/30/60 list.

How does a 22-second script price on Sume?

Sume prices Avatar 1.0 video per second, by quality tier, with a slightly higher rate when a product image is attached. At the list rates on the API pricing page, a 22-second clip multiplies out as follows. The product column applies only when you send product_image; omit it for a productless avatar video.

What to do with a script that does not fit

Sume estimates the spoken duration from the script, or sums the duration values you set on each scene in video_inputs. If the estimate falls outside 4 to 60 seconds, the request is rejected, and the docs tell you to shorten the script or split it across jobs. A multi-scene plan still has to land inside the same window, and the current execution expects one avatar per final video with backgrounds that resolve to one shared scene.

If you are coming from a 15/30/60 menu, there is nothing to round up to. Write the script for the time the idea needs, check the estimate, and keep silence beats explicit: a voice.type of silence needs a duration and no script text.

• Under 4 seconds: add a beat or a second line, since the floor is 4 seconds.
• Over 60 seconds: split into two or more jobs and join them afterwards, for example on the timeline.
• Not sure of the estimate: create an avatar video preview first, which uses the same duration window.

What this comparison does not settle

It does not say which service looks better, and it cannot convert JoggAI credits into dollars from the pages read. It only shows that the two services describe length differently: a menu of three script lengths on one side, a continuous window with per-second rates on the other. If your clips are all 15, 30 or 60 seconds, the difference is small. If your scripts vary, the per-second window is the one that avoids padding.