3:2 photo ratio image API: which Sume models list it and which do not
13 of 16 Sume image models list 3:2 natively; GPT Image 2.5 and Imagen 4 do not. Names, per-image prices, and how GPT reaches 3:2 with custom pixels.

3:2 is the ratio of a full-frame camera sensor and a 4x6 inch print, so it is the ratio people ask for when they want a photo-like image. On Sume you can send aspect_ratio: "3:2" to 13 of the 16 image model families in the catalog. It is not on the list for google/imagen-4-fast, google/imagen-4-ultra, openai/gpt-image-2.5.
The lists come from the per-model native catalogs in Sume's contract package (read 2026-10-07). A model accepts only the values its catalog descriptors list; anything else returns 400 unsupported_parameter rather than being silently snapped.
Models that list 3:2, cheapest first
Prices here are Sume's list-times-1.25 figures. The catalog states the billable formula as "list × 1.25 → ceil usd cents", so treat the dollar amounts as the pre-rounding value and read the exact charge from billable_amount_usd_micros in the submit envelope. Failed or cancelled generations are not billed.
| Model id | Price per image (default tier) |
|---|---|
| higgsfield/soul | $0.005 |
| x-ai/grok-image | $0.025 |
| qwen/qwen-image | $0.025 |
| black-forest-labs/flux.2-pro | $0.0375 |
| bytedance-seed/seedream-5-lite | $0.04375 |
| bytedance-seed/seedream-4.5 | $0.05 |
| recraft/recraft-v4 | $0.05 |
| black-forest-labs/flux.2-flex | $0.0625 |
| ideogram/ideogram-v3 | $0.075 |
| ideogram/ideogram-v4.5 | $0.075 |
| qwen/qwen-image-max | $0.09375 |
| google/nano-banana-2.1 | $0.10 |
| google/nano-banana-pro | $0.1875 |
Getting 3:2 from GPT Image 2.5
The GPT Image 2.5 preset list is auto, 1:1, 16:9, 9:16, 4:3, 3:4, 5:4, 9:8 and 4:5. For 3:2, send custom pixels instead: image_size of 1536x1024 (both edges are multiples of 16, well inside the 3840 cap and the 3:1 limit). That prices at $0.0515 at high, $0.01287 at medium and $0.006 at low.
Imagen 4 Fast and Ultra cannot do it: their catalog is fixed at five ratios and takes no custom pixels. Use 4:3 and crop 6.25 percent off the top and bottom, or pick a model from the table.
Request
Use one sizing field per request.
{
"model": "bytedance-seed/seedream-5-lite",
"prompt": "street photo, rainy tram stop, 35mm",
"aspect_ratio": "3:2"
}Source: Sume Image API docs and GET /v1/images/models/{id}/endpoints for the definitive per-model list (read 2026-10-07).
Sources
Related posts
More in Models
- AI video generator for TikTok: which Sume models make 9:16
Six Sume video models list 9:16 and one never takes an aspect ratio. Table of ratios, seconds and resolutions, read from the catalog on 2026-10-07.
- Anime-style ad with Seedance 2.5: use a character sheet as references
Send your mascot's character sheet as input_references to seedance-2.5 on Sume and test a 6-second anime-style ad at 480p before you pay for 720p.
- ByteDance models on Sume: Seedance family and Seedream prices
Seedance 2.5, 2.0, Fast and Mini plus Seedream 5 Lite, 4.5 Edit and v4, with Sume prices, limits and the pick for each job. Read 2026-10-07.
- Can Gemini Omni Flash speak my script? It takes no audio input on Sume
On Sume, Gemini Omni Flash 1.1 takes image and video references but no audio reference. It makes its own sound. For your script in a face, use avatar video.
Written by Sume