Ideogram Reframe aspect_ratio snaps to 1K; Sume's WxH mapping
Ideogram Reframe with Nano Banana 2 resolves aspect_ratio to the closest 1K size the model supports. Sume maps WxH on Nano Banana to the native aspect_ratio.

On Ideogram's Reframe with Nano Banana 2, aspect_ratio "resolves to the closest dimensions the model supports at the 1K output tier", so the ratio you ask for can be adjusted and the returned width and height are what counts. Sume's Nano Banana path maps a WxH size to the native aspect_ratio and documents exact pixels as a post-step, not a guarantee.
Ideogram facts are from its endpoint reference; Sume facts from Image 1.0 and Image generation, read 2026-10-01.
What does Ideogram's endpoint do?
POST /v2/image/reframe/nano-banana-2 expands an uploaded JPEG, PNG or WEBP (up to 50MB) to a new ratio. It is always asynchronous: the 200 response carries a generation_id, a seed, and the resolved 1K width and height, and you poll GET /v2/generations/{generation_id}. Read the dimensions from that acknowledgement instead of assuming your ratio was honored exactly.
How does Sume treat aspect ratio on Nano Banana?
The docs say that on Nano Banana, WxH maps to the native aspect_ratio (their example: 1080×1350 becomes 4:5), and that exact pixels are a documented post-step via job target_pixels. The route takes aspect_ratio and a resolution tier of 512, 1K, 2K or 4K. Check the listed values for the model you use.
| Question | Ideogram Reframe | Sume image API |
|---|---|---|
| Ratio parameter | aspect_ratio, required | aspect_ratio, optional |
| Snapping | Closest size the model supports at 1K | WxH on Nano Banana maps to the native ratio |
| Size tier | 1K output | resolution: 512, 1K, 2K, 4K |
| Input image | Required upload | Not part of that table row |
What happens if I send an unsupported parameter?
The Sume docs say a parameter the selected model does not list is rejected with 400 unsupported_parameter rather than silently dropped. So a snapped ratio is a documented mapping, while an unlisted field fails loudly.
How do I get the exact output size I need?
Pick the ratio first, then check the returned dimensions, then resize or upscale if a platform needs exact pixels. Upscale to an exact pixel size covers that step, and the ratio changer guide covers the editing side.
Sources
Related posts
More in Developers
- Ideogram remove-background 200 with an error body vs Sume errors
Ideogram's remove-background can return HTTP 200 whose body holds error and status_code. Check the body, not just the status. Sume reports failures on the job.
- Ideogram API seed for repeatable results vs Sume's 400
Ideogram endpoints take an optional seed for repeatable results. Sume's image schema has seed, but no model advertises it, so it returns 400.
- Ideogram Topaz Bloom 2 creative upscale prompt vs Sume upscale
Ideogram's Topaz Bloom 2 upscale takes an optional prompt. Sume's image upscale takes image_url and upscale_factor only, with no prompt field. Request shapes.
- Ideogram Topaz Text Refine 8x limit: 8192px math vs Sume upscale
Ideogram rejects a Topaz Text Refine upscale if the output exceeds 8192px on a side. Work out the largest source per factor; Sume's upscaler takes 1 to 4.
Written by Sume