Grok Imagine publishes a plain one to fifteen
The xAI docs state the allowed duration range as one to fifteen seconds, with no condition on resolution, model variant or input mode. Generated videos also include an audio track by default, which is stated on the same page and changes what a piece arrives as. As of 2026-09-22.
| Property | What the page says | Consequence for a brief |
|---|---|---|
| Allowed range | 1 to 15 seconds | Any whole second inside it can be asked for |
| Resolution condition | None stated | The cheapest tier reaches the full range |
| Audio | Included by default | A piece arrives with sound whether or not it was wanted |
Inclusion rule. Statements on the video generation page that bear on how long a single request runs. Rate limits and quotas are a different subject and are not recorded here. Order. The order a request is assembled in: length, then quality, then what comes back.
1An unconditional range is rarer than it looks
Of the rows in this column that publish a range, most attach something to it. One requires a model identifier past the first step, one withholds its longest value below 1080p, one blocks its longest value at 1080p, and one publishes a range per model version. This row publishes a range and stops.
The practical value is that a single number planned against it stays true. A shot list written in eleven second beats does not have to be re-checked against the resolution the placement ended up wanting, and a cheap draft at 480p is the same length as the final.
2One second is a floor worth noticing
The bottom of the range is unusually low. Most rows here start at four or five seconds or publish a fixed six, and a one second minimum makes short inserts available as their own request rather than as trims from something longer.
For advertising that is mostly useful in assembly. A join between two longer pieces often needs a beat rather than a shot, and generating that beat at its real length costs less than generating five seconds and discarding four of them.
3Audio by default alters what the length contains
A fifteen second request here returns fifteen seconds of picture and fifteen seconds of sound. That is not what the duration parameter appears to promise, and it is the reason this page keeps the audio statement in the same table as the range.
Ad placements this register reads state nothing about whether an audio track must be present, so the default is neither required nor forbidden. It is a deliverable that arrives unasked, and a piece that will be laid under a separate voice track needs its own audio handled rather than assumed absent.
4Where this cell sits in the column
| Generator | Duration |
|---|---|
| Amazon Nova Reel | 6 s increments, up to two minutes |
| CogVideoX | 6 s, at 49 frames |
| Google Veo | 4, 6 or 8 s |
| Grok Imagine | 1 to 15 s |
| Hedra | 10 minutes |
| HeyGen | 30 minutes per scene |
| Kling AI | Not stated |
| LTX-Video | Frames, never seconds |
| MiniMax | 4 to 15 s |
| Pika | 5 s and 10 s, 20-25 s on Pikaframes |
| PixVerse | 1 to 15 s on version 6 |
| Runway | 5 to 15 s on Gen-4.5 |
| SceneMixer | Thirty seconds on the Wan tiers, fifteen on H3 |
| Vidu | Not stated |
| Wan 3.0 | 2 to 30 s |
Inclusion rule. Every generator with a row in this register, carrying whatever its own page states as a length. A page that states none keeps its row and says so. Order. Alphabetical by generator name.
- Duration parameterThe allowed range is 1-15 seconds
- Whether sound comes with the pictureGenerated videos include an audio track by default
5Sources
Read from the xAI video generation docs, checked 2026-09-22. The column itself is described on duration, and the rule that keeps surfaces and generators apart is on method. Next along this column: Hedra, HeyGen.