Are your hard drives crammed with 800GB of different style LoRAs? Want a better solution?

This workflow uses the incredible comfyui-stylebook nodes to let you apply virtually ANY STYLE a model can produce, using pure prompting wizardry instead of heavy LoRAs.

It comes packed with 650+ visual styles, 900+ artists, and modifiers for lighting, era, mood, and more! The secret sauce? It feeds your model rich, written descriptors of the styles and artists, so the look lands perfectly even if the model has no idea who the artist actually is.

🔥 The Best Part? Your mileage may vary depending on the checkpoint, but let me tell you: These nodes are absolute MAGIC with Krea2 and many other models!

Features at a glance:

• Zero dependencies or LoRAs needed. None. Zip.

• 650+ Styles & 900+ Artists seamlessly chained into perfect positive and negative prompts.

• No guessing. Don't know what "risograph" or "wet plate collodion" looks like? Browse the massive visual gallery right here before you generate!

How to use it:

1. Install the missing comfyui-stylebook https://github.com/EnragedAntelope/comfyui-stylebook nodes via ComfyUI Manager.

2. Type in your subject (e.g., "a woman in a red coat waiting at a tram stop").

3. Pick a style, artist, or modifier from the drop-downs (or set them to random!).

4. Hit generate and watch the magic happen.

Special note: this also works amazingly well with Identity Forge for infinite coherent subject/scene/style variety! Reference Civit post, or check it out on GitHub.