Why an AI image costs list × 1.25 on Sume: worked examples

Sume's image catalog prices a model at its list price times 1.25, rounded up to a cent. Worked examples for five models and where the number shows.

4 min readSume
All posts

For most models in Sume's image catalog the price you pay for one image is the model's list price multiplied by 1.25, rounded up to the next cent. In the catalog code that rule reads list × 1.25 → ceil usd cents, and the amount billed to your wallet is returned as usage.cost.

The examples below are computed from Sume's pricing code on 2026-09-29. The Image API docs say the endpoint pricing line already includes Sume's margin, so cost_usd × n is what you pay. GPT Image models are priced from image tokens instead and are not in this table.

What do five models come to?

List price in Sume's catalog, × 1.25, and the Sume estimate for one default image, from Sume's pricing code, 2026-09-29.
ModelList× 1.25Sume estimate
google/nano-banana-2$0.080$0.1000$0.10
black-forest-labs/flux.2-pro$0.030$0.0375$0.04
black-forest-labs/flux.2-flex$0.050$0.0625$0.07
ideogram/ideogram-v3$0.060$0.0750$0.08
recraft/recraft-v4$0.040$0.0500$0.05

Why does the estimate round up?

The catalog's stated formula ends in "ceil usd cents". A $0.0375 result becomes $0.04. Batches multiply the per-image list price by n before that step, so the total for n images is not always n times the single-image estimate.

Does the list price match the vendor's?

Where a vendor publishes a starting price it can be checked. Black Forest Labs lists FLUX.2 [pro] from $0.03 and [flex] from $0.05, the same figures as the list column above. Vendor prices can move; the catalog reports what Sume charges now.

Where do I read the price for my request?

On GET /v1/images/models/{model_id}/endpoints, in the pricing line, and afterwards as usage.cost on the response.

curl "https://api.sume.com/v1/images/models/black-forest-labs/flux.2-pro/endpoints" \
  -H "Authorization: Bearer $SUME_API_KEY"

Sources

Related posts

More in Pricing

All Pricing posts

Written by Sume