Captions horizontal 16:9 AI Edit vs Sume avatar aspect ratios
Captions AI Edit now outputs 16:9. Sume avatar videos accept 16:9 too, at 720p only, with a script or plan estimated at 4-60 seconds. 9:16 is the default.
For a 16:9 talking-head clip in Sume, send aspect_ratio: "16:9" to POST /v1/avatar-1.0/talking-video. Resolution is currently 720p only, and the script or plan must estimate at 4-60 seconds. The default ratio is 9:16.
Captions facts are from its release notes; Sume facts from Generate avatar video, read 2026-10-01.
What changed in Captions?
The release notes say AI Edit now supports horizontal (16:9) video output, useful for YouTube and desktop-first content.
Which ratios can a Sume avatar video use?
aspect_ratio supports 1:1, 3:4, 9:16, 4:3 and 16:9.
| Setting | Value in the docs |
|---|---|
aspect_ratio | 1:1, 3:4, 9:16, 4:3, 16:9 |
| Default ratio | 9:16 |
resolution | 720p (current) |
| Accepted length | 4-60 s estimated |
How do I make the request?
Include avatar_handle, exactly one of script or video_inputs, aspect_ratio: "16:9" and an Idempotency-Key header. Optional quality is standard, plus (default) or max. For a wider scene, describe the background in scene or background prompts. See also all five avatar ratios.
What if my video is longer than 60 seconds?
Shorten the script or split it into multiple jobs; the docs say longer scripts are not accepted.
Sources
Related posts
More in Use cases
- Captions app SRT export vs Sume burned-in caption cues
The Captions app can export an SRT file for external caption tracks. Sume burns captions into the MP4 from authored cues with text, start and end.
- Captions AI avatar looks vs one Sume avatar handle per look
Captions added Avatar Looks to save new looks per avatar. On Sume you create one avatar handle per look from a prompt, a profile or a reference image.
- Multilingual video captions: language is a hint, not a font
On Sume video captions, `language` only tells speech-to-text what to expect. Look and font come from `style`, `design` and `font`, never from the language.
- Chrome Web Store 440x280 promo tile: generate at 1760x1120
The Chrome Web Store small promo tile is 440x280, below GPT Image 2.5's pixel floor. Request 1760x1120 (exact 4x, same 11:7 ratio) on Sume and downscale.
Written by Sume