Nova Reel gives exact pixels and never names a shape
Six of the fifteen entries in this register say nothing about output shape. This one is the reason the column is worth a page anyway: it publishes an exact pixel resolution, which answers the shape question by arithmetic, and still never states a ratio. As of 2026-09-22.
| Question | What the page answers | What this register records |
|---|---|---|
| How large is the output | 1280x720 | Recorded as published |
| What shape is the output | Nothing | Not stated |
| Is a vertical output available | Nothing | Not stated, and not inferable from one pair |
Inclusion rule. What this user guide states against the two questions the shape column asks. An arithmetic consequence of a published figure is not recorded as a published figure. Order. The question the page answers first.
1Why a derived ratio does not go in the cell
1280 by 720 is sixteen by nine. Writing that in the shape column would make this row look like the four entries that publish two or more shapes, and it would be wrong in the way that matters: those pages say a vertical output exists and this one does not.
A single pixel pair says what one output looks like. It cannot say whether the parameter accepts another value, and a placement plan needs the second answer. The hollow cell is the honest record of that gap.
2The consequence is that vertical cannot be planned from this page
Most short-form advertising runs vertical. A page that names one widescreen resolution and no shape parameter gives a buyer nothing to book a vertical placement against, whatever the product may be capable of.
That is a sharper limitation than a missing tier. Resolution can be revisited after a render and shape cannot, so an unanswered shape question blocks the brief rather than the delivery.
3The other five silences are ordinary; this one is not
Two of the other silent rows are priced pages that print resolutions and nothing else. Two are model cards describing research releases. One is a rate card with a 4K tier. In each case the whole page is thin on output specifics, and the shape gap is part of that.
Here the page is otherwise precise: one resolution, one frame rate, a length rule and a model identifier condition. The shape column is the single question it leaves open, which is why the register gives this silence a page and leaves the other five as marked cells in a chart.
4Where this cell sits in the column
| Generator | Aspect ratio |
|---|---|
| Amazon Nova Reel | Not stated |
| CogVideoX | Not stated |
| Google Veo | 16:9 and 9:16 |
| Grok Imagine | Seven, 16:9 the default |
| Hedra | Seven, including 9:21 and 21:9 |
| HeyGen | 16:9 or 9:16 |
| Kling AI | Not stated |
| LTX-Video | Not stated |
| MiniMax | Six, required for text-to-video |
| Pika | Not stated |
| PixVerse | Eight on version 6 |
| Runway | Seven pixel pairs |
| SceneMixer | 9:16 and 16:9 |
| Vidu | Not stated |
| Wan 3.0 | Six, adaptive by default |
Inclusion rule. Every generator in the register, carrying whatever its own page says about output shape. Six rows say nothing and are marked hollow rather than dropped. Order. Alphabetical by generator name.
- Video resolution1280x720the single value in the table
- Video duration6 second increments, up to two minutesfrom the model table
5Sources
Read from the Amazon Nova user guide, checked 2026-09-22. The column itself is described on aspect ratio, and the rule that keeps surfaces and generators apart is on method. Next along this column: TikTok, SceneMixer.