Pick a video model for a Reel by length: 3, 10, 15 or 30 seconds
The length you need narrows the Sume models fast: 2 s is Wan only, 10 s allows Omni, 30 s allows Seedance 2.5 or Wan. Table of the shortlist.
A Reel can run to 3 minutes, as NapoleonCat reports, but each Sume video job is far shorter. Start from the length of one shot and the model list shrinks.
Shortlist by single-job length
Seconds ranges are from the Sume catalog and router limits.
| Job length | Models that accept it |
|---|---|
| 2 s | wan-3.0 |
| 3 s | wan-3.0, gemini-omni-flash-1.1 (sume/auto) |
| 5 s to 10 s | all of the above plus Seedance 2 family, Seedance 2.5, Kling 3, MiniMax H3, Grok |
| 11 s to 15 s | Seedance 2 family, Seedance 2.5, Wan 3.0, Kling 3, MiniMax H3, Grok (image to video) |
| 16 s to 30 s | seedance-2.5, wan-3.0 |
Past 30 seconds
Above 30 seconds you need a join. The even-split planner shows how to size the jobs so none is below a model's minimum. The shortest valid job is covered in the 2-second post.
Sources
Related posts
More in Models
- Portrait 4K AI video on Sume: Omni Flash 9:16 at $0.375 a second
LTX-2.5 is reported to do portrait up to 4K. On Sume the 4K portrait option is Gemini Omni Flash 1.1 at $0.375 a second: 3 s $1.13, 10 s $3.75.
- Qwen Image 2.1 sizes (2752x1536, 2400x1792) vs Sume Qwen ratios
The Qwen-Image-2.1 card lists 2048x2048, 2400x1792 and 2752x1536. Sume qwen-image takes 13 aspect_ratio values instead of pixels. How they line up.
- Qwen Image 2.1 Pro API: the model card lists no Pro variant
Searching for a Qwen Image 2.1 Pro API? The 2.1 model card names no Pro variant and no hosted endpoint. Here is what Sume lists for Qwen today.
- Qwen Image 2.1 Pro: what exists, and which Qwen ids Sume lists
People search for Qwen Image 2.1 Pro. Its model card lists no Pro variant; Sume's Image API lists qwen/qwen-image and qwen/qwen-image-max.
Written by Sume