Docs

AI images

Generate images in the editor with your own OpenAI or OpenAI-compatible image API key.

On this page

AI images adds a Generate with AI button to image fields in the editor. You describe a picture, choose its shape, and the image is created and saved to your media library, ready to use. It uses your own API key for OpenAI or another image service with the same API, so you pay that service directly.

Set it up

You need an API key from OpenAI (opens in a new tab) (API keys page), or from another service that offers an OpenAI-compatible image API.

  1. Open the AI writing tab

    Go to Uncoder → AI & MCP → AI writing and scroll to the AI images card.

  2. Turn on image generation

    Turn on Generate images in the editor.

  3. Paste your key

    Paste the key into API key.

  4. Check the model and endpoint

    For OpenAI, keep the defaults: Model gpt-image-1 and the OpenAI Endpoint. For another service, enter its image model name and the address of its image generation endpoint.

  5. Save

    Click Save changes in the bar at the bottom of the screen.

The AI images card with Generate images in the editor turned on and the API key, Model and Endpoint fields
AI & MCP → AI writing → AI images

Like the writing key, the image key stays on your server: it is never shown again, never sent to the editor and never exported. Leave the field empty to keep the saved key.

Generate an image in the editor

  1. Select an element with an image

    For example an Image widget. In its image field, click the Generate with AI button (the magic wand).

  2. Describe the picture

    In Describe the image, write what you want to see, the style and the light. For example: "A cozy reading nook by a tall window, oak bench with linen cushions, soft afternoon light, photo".

  3. Choose a shape

    Pick Landscape, Square or Portrait.

  4. Generate

    Click Generate. Creating an image can take up to a minute. The result appears in the dialog with the note Saved to the media library.

  5. Use it

    Click Use this image to put it into the field. Not right yet? Change the description and click Generate again.

The Generate an image dialog with a description of a reading nook, the shape buttons, Generate and Use this image
Generate an image from an image field.

Every generated image is saved in your media library, also the ones you did not use. Its description becomes the image's title and alt text. Review the alt text afterwards: a short, plain description is usually better for visitors who use screen readers.

Sizes

ShapeSize in pixels
Landscape1536 × 1024
Square1024 × 1024
Portrait1024 × 1536

Who can use it

Users who can edit with Uncoder and upload files see the Generate with AI button once AI images is set up, unless their role has No access in Roles & access.

Limits and privacy

  • What is sent: only the description you type, when you click Generate.
  • Where it goes: from your server to the endpoint in the settings, with your key.
  • Rate: each user can generate 20 images in 10 minutes.
  • Check the result: generated images can contain mistakes, such as odd hands or garbled text. Look at each image before you publish it.

If generating fails, the dialog shows the reason. "The image service rejected the API key" means the key is wrong or was revoked.