Seedance output size in pixels: what 720p 9:16 or 1080p 21:9 means
Pixel dimensions Sume uses for Seedance by resolution and aspect ratio: 720p 16:9 is 1280x720, 1080p 21:9 is 2205x945, 480p 9:16 is 496x864.
On Sume, Seedance sizes come from the resolution and aspect ratio together: 720p 16:9 is 1280x720, 720p 9:16 is 720x1280, 1080p 16:9 is 1920x1080 and 1080p 21:9 is 2205x945. These are the dimensions the price estimate uses, from the repo pricing tables read 2026-10-05; the API does not accept a size field and returns 400 unsupported_parameter if you send one (Video generation).
What are the sizes?
480p is not 480 pixels on a side: it is a 496-pixel or 864-pixel edge for 9:16, and a narrower box for 21:9. The table lists four common ratios; 4:3 and 3:4 are in the catalog too.
| Resolution | Aspect ratio | Width x height |
|---|---|---|
| 480p | 21:9 | 992x432 |
| 480p | 16:9 | 864x496 |
| 480p | 1:1 | 640x640 |
| 480p | 9:16 | 496x864 |
| 720p | 21:9 | 1470x630 |
| 720p | 16:9 | 1280x720 |
| 720p | 1:1 | 960x960 |
| 720p | 9:16 | 720x1280 |
| 1080p | 21:9 | 2205x945 |
| 1080p | 16:9 | 1920x1080 |
| 1080p | 1:1 | 1440x1440 |
| 1080p | 9:16 | 1080x1920 |
Why does the size matter for price?
Seedance is billed on video tokens: width x height x seconds x 24 fps / 1024. A wider box has more pixels, so more tokens. At 720p, 21:9 is 1470x630 = 926,100 pixels against 921,600 for 16:9, so they cost almost the same; see Seedance 2.5 aspect ratio price.
How do I request a size?
Send resolution and aspect_ratio, not size.
curl -X POST https://api.sume.com/v1/videos \
-H "Authorization: Bearer $SUME_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model":"seedance-2.5","prompt":"A lighthouse at dusk, waves","duration":4,"resolution":"1080p","aspect_ratio":"21:9"}'Does the file I download have exactly those pixels?
Inspect the downloaded MP4 with the video inspect tool if the exact output size matters for an ad spec, rather than assuming the table.
Sources
Related posts
More in Models
- Seedream 4.5 edit with n=4 returns one image on Sume
On Sume, a Seedream 4.5 call with reference images is an edit and always asks the provider for one image, so n=4 does not give four takes. Use four calls.
- Seedream 5.0 Flash takes 14 references; Sume seedream-5-lite takes 10
Seedream 5.0 Flash lists up to 14 input references on OpenRouter. Sume's catalog caps references at 10 and lists seedream-5-lite, not Flash.
- Spain vs Mexico, Brazil vs Portugal: regional voiceover on MAI, Sume
MAI-Voice-2.1 lists es-ES, es-MX, pt-BR and pt-PT. Sume's guard compares only the primary language. How to pick a regional read and test it for 1 cent.
- Text edits on Sume: use sume/auto or pin ideogram/ideogram-v4.5?
Sume's sume/auto never says which model ran, so a text-edit workflow that depends on Ideogram 4.5 behavior should pin ideogram/ideogram-v4.5.
Written by Sume