Qwen Image vs Qwen Image Max for 25 square menu tiles: $0.63 vs $2.34

For 25 square menu tiles, Qwen Image costs $0.625 on Sume and Qwen Image Max $2.34375, a factor of 3.75. Both list 1:1. The two differ in one capability that decides workflows: Qwen Image takes input_references, Qwen Image Max is text-to-image only. The Image API docs explain how the catalog gates parameters.

If your menu tiles start from your own dish photos, Max is out before price matters.

What does each tile cost?

Both are flat per-image rows, so the maths is multiplication. The only variable is how many tiles you regenerate.

What happens if I send a photo to Max?

Sume does not silently drop the field. A request that carries input_references to a text-only model is rejected with 400 unsupported_parameter and a message that the model is text-to-image only. That is useful: you find out on the first call, and you are not billed for a failed request.

Read the model's supported_parameters from GET /v1/images/models before a batch. For a text-only run, the safe pattern is to submit one tile, check the result, then queue the other 24.

Is Max worth 3.75 times the price?

That is a question for your menu. Run the same dish prompt on both for one tile each, look at plate detail, and judge at the size customers see on a phone. Do not expect a published benchmark to settle it. Prices here are list x 1.25 in dollars per image before whole-cent rounding; the catalog's billable formula reads "list x 1.25, ceil usd cents", so confirm the first charge in usage.cost and budget from the invoice, not from the sum.