Log in

Guide to selecting and using Stable Diffusion AI models

How to select AI models in's Text to Image tool

On we provide over 60 image-generation AI models, each specialized and best in producing different results. Selecting the right model is crucial to achieving your creative goals. In this guide, you will learn about:

  • different models categories on

  • using the model selection tool and how to change navigate between models

  • managing your favorite models

The availability of models depends on the tool that you are using.

AI Canvas and Image Editor offer a limited number of models that work with pipelines available in those tools, i.e., outpainting or inpainting.

AI Generator is divided into Essential and Advanced. Essential AI Generator offers three simple styles - Photorealistic, Artistic, and Anime, which, in fact, are our custom-trained models. Advanced AI Generator offers a full spectrum of 60+ models categorized into 6 groups.

How do we categorize models on Advanced AI Generator?

We have divided our models into six categories based on the purpose and results that each model produces:

  • General / Art - the main category for universal models capable of generating anything

  • Photorealism - models best for creating real-life like photos and photorealistic art

  • Styles - includes all models that are generating images in specific styles, like Van Gogh or Disney

  • Anime - providing the best set of anime AI models

  • Design - where you can find models for architects and designers

  • Dreambooth - is the place for all your Dreambooth models

And last but not least, you can find all models and add your favorites in the last two tabs. A search bar on the right lets you easily find what you're looking for.

Choosing a suitable model is an essential step in your creative process. Each model was trained to serve a specific purpose, and the categories above may not cover all use cases for each model, so sometimes, it's a matter of experimenting.

Model selection

Selecting a model in our Text to Image tool is simple and intuitive. If you are working in Essential AI Generator, just click the right button with the style of your choice in the prompt input field.

Advanced AI Generator offers a wider choice of community-trained models. To see the list of all available models, click on the first section of the left-hand side menu of our Text to Image tool.

To choose your model, click on it, and it will be immediately ready to use. You can then go ahead and write your prompt and adjust different parameters.

Most models on are open-source, and community trained. We are crediting the authors o each model. When hovering over the model preview, you can find the name and the url there. Please take a time to support them.

While hovering over a model, an option to add it to favorites will appear.

Managing your favorite models

Hover your mouse over a given model, and you will see the 'heart' on the top left side. When you click on the 'heart' icon, it will turn red, and the model will be automatically added to the Favourite menu, which allows you to jump to the models you frequently use quickly.

Comparing some of the models

analog style, close portrait of a beautiful woman, photorealistic, intricate hair, art by Diego Fazio, diego koi, oscar Ukono, concept art, sharp focus, art gem, 8k highly detailed, shoot with DSLR

SynthwavePunk v2
Analog Diffusion

rusty warship dreadnought shipwreck in a lush forestsci-fi, ivy, moss, trending on artstation, concept art by Greg Rutkowski and Sebastian Luca, global illumination

ICBINP Afterburn

tree but minimalistic 3d concept art by frank stellagilleard james, whalen tom, colorful, octane render, vray, depth of field, trending on artstation, minimalism

Vectorartz Diffusion
Van Gogh Diffusion

We aim to provide the best AI models and enable fast generation with premium GPUs. Our model library is constantly growing. If you have any suggestions for the new additions, feel free to add them to our feature requests here or contact us directly.

The most actual list of models is available in our application in the model selection menu.