AI image generation is billed per image, and the settings that drive the price are the ones people change last. Setting them first is the difference between a few cents and a few dollars across a page build.
This page covers the settings that affect what each generation costs you.
Before You Start
- The AI image generator already set up. See the AI image generator doc first.
- You have to be an administrator. Measured across four roles: the AI endpoints require the
manage_optionscapability, which administrators have and editors, authors and subscribers do not. An editor cannot generate AI content.
Configure Image Settings
- The model is the biggest lever. Measured, the default is
dall-e-2withdall-e-3selectable; the newer model costs more per image. - Size and quality are priced by the provider, not by Nexter. A larger image is a more expensive request.
- The number of images per request multiplies the cost. Asking for four variations is four images.
- Every follow-up is another billed request. Editing, expanding, resizing, generating variations and enhancing a prompt are each their own call.
How to Keep The Cost Down
- Draft on the cheaper model, finish on the better one. Get the composition right before switching.
- Generate one image at a time while you are still iterating.
- Write the prompt properly before generating, rather than generating and editing repeatedly — each edit is a fresh charge.
- Generate at the size you will actually use. A hero image and a thumbnail do not need the same request.
Limitations
- What this page verified, and what it did not. Setup, permissions, storage, provider selection and the exact failure messages were measured on a site with no API key. Generation itself was not exercised, so nothing here describes output quality, token cost or rate limits.
- There is no cost display in the editor. Nothing shows a running total or warns you before an expensive request; your provider’s dashboard is the only place the spend appears.
- Nexter does not set the prices and cannot cap them. Use your provider’s own spending limits if you need a hard ceiling.
- Cheaper settings produce weaker images. This is a trade, not a free saving.
- Model pricing changes on the provider’s schedule, not this plugin’s.
Troubleshooting
My bill is higher than expected
Count the operations rather than the images kept. Edits, expansions, variations and prompt enhancements are each billed, including the ones you discarded.
Can I set a spending limit here?
No. Set it in your OpenAI or Google account; this plugin has no cap of its own.
Which model should I use?
Draft on the cheaper one and switch to the better one for the final image. The gap in quality is smallest when the prompt is already good.
