Cover image for Harnessing the Power of Presets in Fooocus.

Posted on • Updated on

Harnessing the Power of Presets in Fooocus.

Image description

Even though Fooocus aims to be user-friendly, its creators understand that with Stable Diffusion, efficiency and customization are not just advantages, they are essential.

Presets, or predefined configurations, embody this spirit by allowing users to tailor the tool to their unique needs. Whether it's to save time by avoiding constantly readjusting settings or to easily switch between different configurations for specific projects, presets are the key to a smooth and tailored workflow.

Whether you're a seasoned user looking to further optimize your creative process or a newcomer eager to understand how to better use Fooocus, this tutorial is for you.

What are presets in Fooocus?

Presets in Fooocus are sets of predefined parameters used to configure Fooocus when you launch it. They are designed to save you time and energy by avoiding the need to manually adjust settings every time you want to generate an image in a particular style.

It's also a simple way to choose the most optimized parameters for the desired generation type or chosen model.

In practice, when you launch Fooocus, you can specify a preset that will set the default parameters for your generations.

These parameters can range from default prompts to models and LoRA to use, as well as image dimensions and the number of inferences used for generation.

Fooocus comes with a series of presets provided by default with the application, and you can also create your own.

Presets provided with Fooocus

The source code of Fooocus provides 5 presets that are directly usable once it is installed:

default: The default configuration used when you do not specify a preset at launch.

anime: An optimized configuration for generating Anime or Manga images using the animaPencilXL_v100 model.

realistic: A configuration chosen for creating portraits and photorealistic images using the realisticStockPhoto_v20 model complemented by the LoRA SDXL Film Photography Style.

sai: The basic configuration recommended by Stability AI for using the basic version of SDXL and its Refiner.

lcm: An optimized configuration for ultra-fast generation with the LCM technique and the same model as the default parameters, JuggernautXL V8.

Top comments (0)