Learn how to generate images in Aymo: write a prompt, pick a model, style, and aspect ratio, then remove backgrounds, upscale, expand, or relight results.

The AI Image Generator creates images from text prompts. Write a prompt, choose a model and settings, and generate. You can edit the result in the same place without generating again.

Plain sentences work. No special syntax is needed. Include the details that change how the image looks:
Example: "A ceramic coffee mug on a wooden table by a rainy window, soft morning light, product photography, close-up with space on the left for text."
Put the most important detail first. Long prompts with many adjectives tend to dilute the result.
Aymo includes multiple image models, such as GPT Image 2, Nano Banana 2, Seedream 4.5, and FLUX.2. Each model handles photorealism, illustration, and text inside images differently.

To find the right model for a task, send the same prompt to more than one model and compare the results. The full list is in the Model Directory.

Results usually appear within a few seconds. Before changing anything, check:
If the subject or composition is wrong, revise the prompt. If only one detail is wrong, use the editing tools instead of generating again.
Download any result to save it. Images generated in Aymo can be used for personal and commercial projects.
To create a consistent set of images, reuse the same model, style, aspect ratio, and prompt structure.
Editing tools change part of an image without regenerating the whole thing.

Remove Background cuts out the subject with clean edges. Useful for product photos, stickers, and layouts.

Upscale increases resolution while keeping details sharp. Generate at a standard size while iterating, then upscale the final image for print or large displays.

Expand extends the image beyond its original frame and fills in the new area to match the scene. Use it to turn a square image into a wide banner.

Adjust Lighting brightens, dims, or changes the mood of an image without generating a new one.
Image models are billed per image. Available models, resolution, and variants depend on your plan. See Billing and Plans for details.