Static ad aspect ratios an image API can return: Sume's list
Which aspect ratios Sume's Image API accepts for static ads, why 4:5 is not 4:3, and why you read the model's own list before pinning one.
Sume's Image API normalizes aspect ratio to 1:1, 16:9, 9:16, 4:3, 3:4, 3:2, 2:3, 4:5, 5:4, 1:2, 2:1, 1:4, 4:1, 1:8, 8:1, 9:21, and 21:9, or auto to let the provider choose. A given model only accepts the subset its catalog lists, so read supported_parameters before you pin one. Static ad batches such as Higgsfield Ads Studio's, which its changelog dated 2026-09-30 says run to 20 ads per product, need one ratio per batch, set deliberately.
Facts here come from Sume's Image API docs, read 2026-10-02.
Which ratios does the Image API accept?
The list above is the normalized vocabulary. It is not a promise that every model takes every value: a request that sets a value the model does not list is rejected with 400 unsupported_parameter. The docs example for Seedream 4.5 shows 1:1, 16:9, 9:16, 4:3, and 3:4 only.
| Ratio | Note from the docs |
|---|---|
1:1 | In the normalized list and the Seedream 4.5 example |
4:5 | Instagram portrait (1080x1350), not 4:3; exact pixels need a post-step |
9:16 | In the normalized list and the Seedream 4.5 example |
auto | Provider choice; on edits, prefer it to match the reference |
What is special about 4:5?
The docs say 4:5 is Instagram portrait at 1080x1350, not 4:3, and that Banana Pro sends aspect_ratio: "4:5" with a native size of about 928x1152 at 1K. Exact 1080x1350 is a documented post-step through the job's target_pixels. So a model can match the ratio and still not hand you the final pixel size.
What about edits from a product photo?
On edit and image-to-image calls, the docs say to prefer aspect_ratio: "auto" to match the reference, and that omitting the field is not the same as auto. If your ad template needs a fixed ratio, set it explicitly and check the result.
How do I check a model first?
Call GET /v1/images/models and read each model's aspect_ratio enum. Then run one image before a batch. 20 ad variations from one product image uses a fixed 1:1 ratio for that reason.
If one creative must ship in several shapes, generate each shape as its own request rather than cropping one image, since a crop can cut the product or the headline. Record the ratio in the file name. For shapes outside a model's list, choose the nearest listed ratio and finish with a post-step, and view the final file at its real size before it goes out.
Sources
Related posts
More in Media tools
- Steam trailer thumbnail: a 1920x1080 frame from the video itself
Steam wants the trailer poster as a 1920x1080 JPG or PNG taken from the video. Pull it with Sume video frames at max_edge 1920, plus the trailer specs to check.
- Subtitle cues from STT sentence segments and the 70 ms lead
Ask Sume STT for sentence segments and tune boundary_lead_ms (default 70) so each cue holds a little past its last word. Settings and limits.
- Telegram video emoji 100x100: Sume's 256 px floor and a square crop
Telegram video emoji must be exactly 100x100. Sume video trim outputs 256 px at minimum, so crop a square and cut in Sume, then downscale and encode outside.
- Telegram video sticker: 512 px, 3 s, 256 KB, WEBM VP9 with Sume
Telegram video stickers need a 512 px side, 3 s, 30 fps, 256 KB, no audio, WEBM VP9. Sume trims and resizes but outputs MP4, so one step stays outside.
Written by Sume