Documentation

Generate

Turn a written prompt into a finished image, with reference images and background jobs.

Generate is the straightforward path: describe the image, choose how it should be made, and run it.

  1. Write your prompt. Plain language works; you do not need special syntax.
  2. Pick a model, then set its aspect ratio and resolution, and choose how many images to make in one go. See models and options.
  3. Optionally attach reference images, either by uploading a file or by dragging one in from the Library.
  4. Run it.

Generations run in the background with live progress, so you can keep working while they finish, and every result is written to your Library automatically. How many run at once is up to you, in Settings.

Happy with a prompt? Save it, settings and all, so you can reach for it again from Prompts.