Sign In

Style Strength Controller [IL]

66
485
86
23
Updated: May 2, 2025
style
Verified:
SafeTensor
Type
LoRA
Stats
343
54
8
Reviews
Published
Apr 29, 2025
Base Model
Illustrious
Usage Tips
Strength: 0.2
Hash
AutoV2
DD70E73F5C

Style Strength Controller

Now you can control overall style strength, even if you are a user and cannot control the training process.

Most importantly, the effect of this LoRA is predictable, linear, and has ZERO side effects on style. (Not an exaggeration, it's really mathematically zero).


NOTE: Forget about v0.1. It is alpha version.

Use v1. It's much better and stabler. Supports all tools. All info in this page has been updated for v1.


What can this LoRA do?

This LoRA can reduce the style of your base model. Or amplify it if strength < 0 and you want something crazy.

This lora will not bring any side effects to the style.

I like the style, why would I reduce its strength?

The real purpose of this LoRA is to reduce overfitting effects and bring creativity back, by just reducing the style a little bit.

Overfitting effects?

Because model was trained too hard, and the dataset has bias.

E.g.:

  • noticeable bias. E.g.: Always too bright/dark, generating same faces / things / backgrounds.

  • too sensitive to some prompt words.

What's the effect of this LoRA?

Here is a test on Hassaku XL v2.1fix: Notice that

  • This base model has a noticeable bias towards high brightness.

  • So at strength -0.3, the model completely ignores the prompt word "dark". Because you amplified the style and bias as well.

  • At strength 0.25. Model has much less bias on brightness and feels more natural. The style doesn't noticeable change.

  • Strength 0.5 is for reference, less style, more natural.


How to use?

  • Just apply it as normal LoRA.

  • Working strength is around -0.5~1. Recommended -0.25 (go crazy) or 0.25 (go stable).

  • You don't have to set the patch strength for text encoder. This LoRA does not patch it.

  • Some styles heavily affect CFG scale. So you may also need to adjust the CFG scales because the style strength changed.


What's the training data? Why it has zero side effect on style?

This LoRA is "calculated/calibrated" directly from SDXL and Illustrious v0.1. No training process.

Supported models?

This LoRA only supports models that are close to the base model it was calibrated for.

Currently there is only one version for Illustrious v0.1. I'm still testing this idea. Other versions, like NoobAI etc. will come later.


Guys, remember to leave feedback in comment section. Don't write feedback in Civitai review system, it was poorly designed, literally nobody can see the review.

Have fun.


Old info for v0.1:

  • Only support ComfyUI.

  • Strength: ~0.5~0.5. >0 will amplify the style, <0 will reduce it.