Skip to content
On this page

Creating styles

How do I create a style?

INFO

To understand terms such as class, prompt, token, and how to add new styles on admin panel, please refer to the Generations page. Also, use Useful links page for more information about Stable Difussion itself and links to websites with style prompts.

OCULA AI uses the Stable Diffusion's algorithm to implement styles on generations. To create your own style, you need to provide prompt and negative prompt texts, set them in the Styles section with a preview image and style name, and specify the number of iterations and the number of images to generate in the Generations Settings section.

Definitions

  • The Prompt is the text that guides the image's generation. It should include a token name that references the subject, which in OCULA AI is [token].
  • The Negative Prompt specifies what should not be included in the image. For instance, if we want to generate an image with a cloudy sky, we enter "clear sky" as the negative prompt.
  • [token] - A unique identifier to represent a specific subject, such as a person or item featured in an uploaded image

How to test new styles?

The video below shows how to create a new style and test it using Google Colab: