Stabilizer

Name is misleading. This is NOT a style LoRA, nor a LoRA that can magically fix your overfitted merged base model.

This is a finetuned model from pretrained model. But trained as a LoRA.

• Diverse dataset. Relatively big enough, 11k images. Not overfitted. No bias, no default style. Still maintains maximum creativity.

• Dataset only contains images with the highest quality. Best textures, lighting and art details.

• Zero smooth plastic glossy AI image in dataset. Glossy Al images are polluting the world, but not on my watch, I handpicked every single image.

• Natural language captions from Gemini, rather than tags in random order with high FPR.

• Standard noise scheduler. Inpainting/img2img friendly.

Why no default style?

If a model was overfitted and has a "default style" (bias), it means no matter what you prompted, the model must generate the same things (faces, backgrounds, feelings) that make up the default style.

• Pros: Easy for beginners. You won't have to prompt style and you will always get a stable style.

• Cons: You can't overwrite the default style, it was overfitted. If you prompt something that does not fit the default style, the model will just ignore it. If you stack more styles, the default style will always overlap/pollute other styles.

"no default style" means no bias. The model will follow your prompt and you must prompt the style you want. You can get the style exactly as it should be (as long as the model knows).

I personally use this model as a clean starting point to stack more style LoRAs.

Improvements:

comparisons with pretrained base model:

• https://civitai-proxy.pages.dev/images/84145167 (general styles, you get the style exactly as it should be)

• https://civitai-proxy.pages.dev/images/84256995 (artist styles, no style shifting because this model does not have bias)

See more xy plots in cover images. Cover images are raw output at 1MP resolution. Styles are from the base model, triggered by prompt

Why LoRA?

• This LoRA is a DoRA (from Nvidia), which is more efficient than traditional LoRA.

• Enough for thousands of training images.

• What? You prefer to download and store a 7GiB checkpoint than a 80Mib LoRA?

Share merges using this model is prohibited. FYI, there are hidden trigger words to print invisible watermark. I coded the watermark and detector myself. I don't want to use it, but I can.

This model only published on Civitiai and TensorArt. If you see "me" and this sentence in other platforms, all those are fake and the platform you are using is a pirate platform.

How to use

Just use the pretrained base model and load this LoRA with strength 1. This is a finetuned base model but trained as LoRA. So "pretrained base model weights" + "LoRA patch weights" = "finetuned base model weights".

Versions:

• cknb (based on ChenkinNoob-XL).

• nbvp10 (based on NoobAI v-pred v1.0). FYI: you don't need CFG hacks (RescaleCFG etc.).

• nbep10 (based on NoobAI eps v1.0).

• illus01 (based on Illustrious v0.1).

Load this LoRA first

This LoRA uses a new arch called DoRA from Nvidia, more efficient than traditional LoRA. But the patch weight is dynamically calculated based on the currently loaded base model weights (which will be changed when you loading other LoRAs). To avoid unexpected changes, load this LoRA first.

Specify styles in prompt

This model does not have an strong default style and is very creative. You must specify the style you want in the prompt.

Merge it to other finetuned models:

I personally disagree this. This is not the model's original intention. But this is a LoRA after all.

Be aware:

This LoRA can't remove glossy shiny plastic AI style. Although the dataset is "AI image free". It can't make AI image polluted 1girl overfitted 50 versions of Nova furry 3D anime WAI or whatever look better. What the model learned is learned. Use a pretrained base model if you want to get rid of overfitted AI style.

What is "overfitted AI style"? This is what Craft Lawrence (from spice and wolf) should be, if you've seen the anime: img. This is what those AI style polluted 1girl overfitted model generated: img

Some base models already merged this model. If you got deformed images with this LoRA even at low strength (e.g. <0.5). Your base model has already merged this LoRA (and you merged it twice). And the model weights got multiplied (because how the DoRA works) and collapsed.

Beware of fake base model creators, aka. thieves. Some "creators" never do the training, they only grab other people's models, merge them, wipe all metadata and credits, and sell it as their own base model.

Update log

Moved here