Skip to main content
Two parameters control how an image is generated: resolution and seed. Everything else is handled by the model, LoRA, Style, and Action.

Resolution

Resolution sets the output dimensions. Pick based on your subject and intended use.
RatioBest for
1:1Gallery display, square compositions
7:9Portraits, single characters
9:7Wide scenes, environments
5:12Full-body shots, tall compositions
12:5Ultra-wide, cinematic crops
13:19Mobile viewing, editorial
19:13Landscape, desktop wallpaper
4:7Tall portrait, close-up
7:4Wide portrait, horizontal
If you plan to animate the result, check the resolution requirements before generating. See Image-to-Video.

Seed

A seed is a number that fixes the random starting point of a generation. Same prompt, same model, same seed — same result every time. Seed is hidden by default. To access it, open the Advanced toggle in the generator. When to use it:
  • You got a result you like and want to iterate without losing the base composition
  • You’re testing how a prompt change affects output while keeping everything else constant
  • You want to reproduce a specific result later
When to ignore it:
  • You’re exploring — leave seed off and let the generator vary freely
  • Most users never need to set it manually

Prompt

The prompt field accepts plain text. Write what you want to see: subject, action, setting, lighting, style. Order matters — put the most important elements first. Prompt Enhancer is on by default and expands short prompts automatically. You can toggle it off in the settings panel if you want the output to follow your exact wording. For a full breakdown of prompt structure, see Prompting Basics.

Prompting Basics

How to structure a prompt that gets consistent results.

Image-to-Video

Resolution requirements for animation.