The app register as a data file
All 252 readings, ten features for each of 25 apps, are in one CSV under CC BY 4.0. A feature missing from an app's pages still gets a row, left blank, so gaps can be counted. As of 2026-09-25.
| Column | Holds |
|---|---|
| fact_id | App, feature and a counter |
| value | The reading in a line |
| applies_to | The part of the page it comes from |
| source_url | The page |
| source_name | The app's page |
| value_since | The day read |
| verified | The day read |
Inclusion rule. Every column in the file. Order. File order.
1Reading the ids
An id like opus-watermark is OpusClip's reading on watermarks; revid-generate-2 is a second reading on Revid's AI generation, here its avatars.
No row holds a price. Where a plan decides a reading, the value names the plan instead.
Apps rename plans and move features between them often. Treat each row as the page on the day read.
The applies_to column names the part of the page a reading comes from: a plan card, a comparison table, a menu or an FAQ answer. Readings from menus are the thinnest evidence and are marked that way.
The totals quoted around the site, such as the nineteen apps that list captions, are simple counts of these rows.
Rows cover 25 apps of different kinds, from clip makers to phone apps, so a filter on the kind of app is often the first step before comparing.
Two apps are read partly from help-centre articles rather than pricing pages: VEED, for clipping and posting, and CapCut, whose tool pages stand in for a pricing page that could not be read.
Quotes are copied from the visible text of each page. Where a quote is a single label from a table or menu, the value beside it says what the label means.
Download short-video-app-ai-features.csv
2Index
3Making the drama before the edit
Upstream of an app feature like “Data”, SceneMixer describes its own generation steps:
SceneMixer writes dialogue into the script analysis, assigns each line to a shot with a speaker and a short delivery note, and the video model performs it — picture and audio generated together, so the mouth forms that sentence in the language you chose.
SceneMixer, AI Dialogue Generator (read 2026-10-13)
16:9 landscape (SceneMixer's default) and 9:16 vertical (the usual short-drama frame), chosen per project.
SceneMixer, Seedance 2.0 in a Real Short-Drama Pipeline — “Which aspect ratios does the pipeline generate?” (read 2026-10-13)