FusionAI
Docs navigation

Getting started

Create

Create

Generate an image

How image generation works inside the FusionAI workspace, what it needs, and how to keep going from a result.

Updated September 8, 2026

Image generation turns a written description into a picture inside your FusionAI workspace. This page covers how it works in practice and how to keep the work moving from a result.

For a broader overview of the capability, see the AI Image Generator page. To start, you will need to be signed in — image generation runs inside the workspace and depends on your configured image capability.

Open the image workspace

Go to the image workspace. This is where you describe an image and generate it. If you are not signed in yet, you will be asked to sign in first.

Describe what you want

The prompt is a set of choices you can check against the result. A useful prompt usually covers:

  • Subject — what should appear, and what matters about it.
  • Composition — how it is framed (close-up, wide, centered, negative space).
  • Lighting — the light itself, described rather than labelled.
  • Material and texture — surfaces you want to read clearly.
  • Environment — where the subject sits.
  • Mood and constraints — the feeling, and what to avoid.

Vague requests produce vague images. For a full method, read how to write better prompts for AI image generation.

Choose an aspect ratio

You pick an aspect ratio for the request — for example square, portrait, or wide. Treat the aspect ratio as the reliable control: the exact output dimensions depend on the configured image provider, so check the actual size of a result before using it. The free resize tool can prepare a chosen result for a specific placement.

Review and refine

Generation is rarely one-and-done. Compare the result against your prompt, change the part that is not working, and try again. Small, deliberate changes are easier to judge than rewriting everything at once.

Keep going from a result

An image is usually an ingredient rather than the finish line. From a result you can:

  • Continue editing it into a new version, when an editing capability is available to you.
  • Take it into another workflow, such as turning a still into a short video — see how to turn an AI image into video.

What is available beyond generation depends on the capabilities configured for your workspace.

Where your images live

Results are kept in your private Gallery — your own Library, not a public page. You can revisit or continue from anything you have made there.

Next steps