Grok Imagine video length: 1-15 s on xAI, 4-15 s on Sume
xAI allows 1 to 15 seconds for grok-imagine-video-1.5. Sume's catalog row allows 4 to 15, so a 2-second clip needs another id such as wan-3.0.

Grok Imagine video length is 1 to 15 seconds in xAI's docs, and 4 to 15 seconds on the Sume Grok row (duration_seconds: { min: 4, max: 15 }). A 1, 2 or 3 second clip is outside the Sume row.
xAI's range is quoted from its video generation docs, read 2026-09-30. Sume's ranges are from its catalog and Video generation docs.
What are the two ranges?
xAI's page says "The allowed range is 1-15 seconds". Sume's row sets a minimum of 4. Both share the 15-second maximum, which is also the cap for every Sume catalog model not listed as longer.
| Where | Seconds |
|---|---|
| xAI, Grok | 1-15 |
| Sume, Grok | 4-15 |
Sume, wan-3.0 | 2-30 |
Sume, seedance-2.5 | 4-30 |
Sume, gemini-omni-flash-1.1 | 3-10 |
Where can I get a 2-second clip on Sume?
The catalog lists wan-3.0 at duration_seconds: { min: 2, max: 30 }, and the docs say it accepts 2-30 seconds. See Wan 3.0 2-second clips. gemini-omni-flash-1.1 starts at 3 seconds.
What happens if I ask for less than 4 on the Grok row?
The row's advertised minimum is 4, so use the catalog value rather than the xAI value. Read capabilities from the models endpoint instead of copying limits between providers.
What should I do?
Keep one duration table per id in your code, filled from the catalog. If you trim clips later, request the shortest advertised length and cut in an editor or Timeline step.
Sources
Related posts
More in Developers
- Image 1.0 input_urls, n and format: deprecated names to replace
Sume's Image 1.0 still accepts input_urls, n and format, but prefers image_urls, num_images and output_format. What each maps to and their limits.
- Image 1.0: text, reference or masked edit - which fields to send
Image 1.0 uses prompt only for text-to-image, prompt plus image_urls for edits or references, and adds mask_image_url for masked edits. Fields and URL rules.
- mask_image_url or mask_url? Masked edits on Sume's image APIs
Image 1.0 takes mask_image_url with image_urls; POST /v1/images takes mask_url with input_references for GPT Image 2.5. Field names, models and limits.
- Image job metadata on Sume: stored on the job, not sent upstream
The metadata field on Image 1.0 and POST /v1/images is stored on the job and not sent to the provider. Use it to tie jobs to your own records.
Written by Sume