4:5, 9:16 and 16:9 together: 15 of 19 Sume image models list all three
Which Sume image models list 4:5, 9:16 and 16:9 in one catalog row, which four miss 4:5, and what each costs per image.

Fifteen of the 19 Sume image models list 4:5, 9:16 and 16:9 together, so one model can cover an Instagram feed post, a Story and a YouTube thumbnail. The four that miss 4:5 are Higgsfield Soul, Grok Imagine, Imagen 4 Fast and Imagen 4 Ultra. Every model lists 9:16 and 16:9.
The 15 that cover all three
Pass aspect_ratio in the request, for example 4:5 for a portrait feed post. 4:5 is Instagram portrait (1080x1350). Nano Banana Pro sends 4:5 natively at about 928x1152 at 1K, and exact 1080x1350 comes from a documented post-step through job target_pixels.
| Model | Billed per image |
|---|---|
| Qwen Image | 3 cents |
| Flux 2 Pro | 4 cents |
| Seedream 4.0 | 4 cents |
| Recraft V4 | 5 cents |
| Seedream 4.5 | 5 cents |
| Seedream 5.0 Lite | 5 cents |
| Flux 2 Flex | 7 cents |
| ChatGPT Image 2.5 | 7 cents |
| ChatGPT Image 2.5 Sunburst | 7 cents |
| Ideogram V3 | 8 cents |
| Ideogram 4.5 | 8 cents |
| Nano Banana 2.1 | 10 cents |
| Qwen Image Max | 10 cents |
| Nano Banana Pro | 19 cents |
| ChatGPT Image 2 | 27 cents |
The four that miss 4:5
These models list 9:16 and 16:9, so a vertical Story or a wide thumbnail works, but a 4:5 request returns 400 unsupported_parameter. The nearest substitute is 3:4, which Imagen lists, or 2:3, which Soul and Grok list.
Grok Imagine adds the phone ratios 9:19.5 and 9:20, which no other model lists.
| Model | Billed per image | Nearest portrait ratios |
|---|---|---|
| Soul (Higgsfield) | 1 cents | 9:16, 3:4, 2:3 |
| Grok Imagine | 3 cents | 2:3, 3:4, 9:16, 9:19.5, 9:20 |
| Imagen 4 Fast | 3 cents | 9:16, 3:4 |
| Imagen 4 Ultra | 8 cents | 9:16, 3:4 |
Practical pick
If one model has to serve three placements, the cheapest full-coverage row is Qwen Image, which lists all three and bills 3 cents. At 5 cents Seedream 5.0 Lite adds image editing. Set the ratio per request, never rely on a crop, and confirm with the live descriptor from GET /v1/images/models.
Edge cases to know
Ratio support is per model, and a request that sets a value the model does not list is rejected with 400 unsupported_parameter rather than adjusted. On edit calls, aspect_ratio: auto matches the reference image, and leaving the field out is not equivalent. Always read supported_parameters from the catalog before you pin a tier or ratio.
For a three-placement campaign, one prompt at three ratios costs 9 cents on Qwen Image, 12 on Flux 2 Pro, 15 on Seedream 5.0 Lite and 21 on GPT Image 2.5 at high. A set of 50 products is therefore $4.50, $6.00, $7.50 or $10.50. Pick the row, then pin the three ratios in code so a copy-paste never sends a ratio the row does not list.
Sources
Related posts
More in Models
- 8:1 strips and 21:9 banners: which image models take them
Nano Banana 2.1 lists 8:1, 4:1, 1:4 and 1:8 on Sume; Pro stops at 21:9 and 9:16. Prices per image for each ratio choice, read 2026-10-08.
- Agent Completion or Format run: where can you pick Haiku 5.5?
Only a Format run takes a catalog model id. An Agent Completion accepts model sume-agent and nothing else, so Haiku 5.5 is a Format-run choice.
- background: transparent on the Sume image API: two GPT rows only
Transparent PNGs from the Sume image API need background: transparent, which only the two GPT Image 2.5 rows list. See the request, price and the 400 elsewhere.
- Can I show HunyuanVideo clips to EU viewers? Section 5(c) read
Tencent's HunyuanVideo license says do not use or display Outputs outside the Territory, which excludes the EU, UK and South Korea. Clause by clause.
Written by Sume