Amazon A+ four-column images: 300x225 minimum, made as 4:3

Amazon's Premium four-column A+ module needs images of at least 300w x 225h. Make them as 4:3 stills at 1024x768, which clears Sume's pixel floor.

4 min readSume
All posts

Amazon's PremiumFourColumnImagesModule takes 3 to 4 columns and each column image must be at least 300 wide by 225 high, which is a 4:3 shape. Ask Sume for 4:3 at 1024x768: it is on the 16-pixel grid and has 786,432 pixels, above the 655,360 floor for custom sizes, so a literal 304x224 request would be refused.

Module limits are from Amazon's A+ Premium modules page and the Sep 30, 2026 release notes; image parameters are from Sume's Image API docs, read 2026-10-01.

What does the four-column module require?

Amazon describes four images in a row under one headline, each with its own subheadline and body text; on mobile the columns become a two-by-two grid. The page lists the columns array as 3 to 4 items.

Four-column module fields, Amazon A+ Premium modules page, read 2026-10-01.
FieldConstraint on the page
columns3 to 4 items, position 1 to 4, unique
imageRequired, min 300w x 225h px
Column headlineOptional, max 30 chars
Column description1 to 5 items, total at most 150 chars
Module headlineOptional, max 80 chars

Why not request 304x224 from Sume?

Sume's custom image_size for GPT models needs both edges to be multiples of 16, a longest edge of at most 3840, an aspect ratio of at most 3:1, and 655,360 to 8,294,400 total pixels. A 304x224 image has 68,096 pixels, so it falls under the floor. Amazon states a minimum, not a maximum, in this table, so a larger 4:3 image still meets the stated rule; confirm how Seller Central crops it before you ship.

Aspect ratio is simpler: 4:3 is in the normalized ratio list the Image API accepts, and 300x225 reduces to exactly 4:3.

curl -X POST https://api.sume.com/v1/images \
  -H "Authorization: Bearer $SUME_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: aplus-four-col-001" \
  -d '{
    "model": "openai/gpt-image-2.5",
    "prompt": "Product on a clean studio surface, three-quarter angle, soft daylight",
    "image_size": "1024x768",
    "n": 1
  }'

Can one call make all four column images?

Only as four takes of one prompt. n requests several images per call: the docs allow up to 10 and say per-model ceilings are lower, with the catalog example showing a range of 1 to 4. Four columns that show four different angles are four prompts, so send four requests, each with its own Idempotency-Key so a retry cannot bill twice.

If you do need a larger batch across SKUs, see bulk image generation API.

What should I check before uploading to Amazon?

Amazon's media flow, from the release notes, adds createMedia, getMedia and updateMedia for A+ media assets. Keep the four files at the same ratio so the grid looks even, and write the short headline and description text separately; those limits are character counts, not something an image API produces. Amazon's page is the authority for the module rules. If a source image has the wrong shape, see the aspect ratio changer.

Sources

Related posts

More in Use cases

All Use cases posts

Written by Sume