models/Control LoRA Collection - cfg scale BOOST

Control LoRA Collection - cfg scale BOOST

7/6/2025

5:31:10 AM

Related Keywords & Tags

advokat,anime,cfg scale boost,cfg scale boost lora,control lora collection,detail lora,image generation,lora,photorealistic,realism lora,sdxl,sdxl 1.0,stable diffusion,styles,tool

A green frog character wearing a patterned poncho squats on grass surrounded by flowers and lily pads with an abstract pink and yellow background.

Cyberpunk boy with black hair, mechanical arm, and tattoos wearing futuristic gloves against a neon-lit city background with archaic Japanese text.

A young woman with pale skin and a hooded cloak stands in a dimly lit gothic corridor, holding an ornate box filled with golden skulls, wearing a gold skull brooch.

Close-up low-angle view of a woman with blue and white streaked short hair, wearing an orange crop top, fishnet underlayer, black baggy pants, and black Converse shoes in a vibrant, neon-lit urban setting.

A black-haired female warrior with big grey eyes and pointy ears, holding a staff, featuring traditional Japanese tattoo designs and wearing a dark leather outfit.

Close-up digital art of a girl with purple eyes, black hair with pink highlights styled in buns with hair horns, spiked choker, fingerless gloves, aqua nails, and a seductive smile.

A luminous female figure with complex iridescent horns, wearing a shiny holographic outfit, illuminated by volumetric lighting against a dark background.

A female yakuza with black hair tied up, exposed cybernetic spine, dragon tattoo on her back, smoking a cigarette in the rain on an ancient Japanese street holding a cybernetic katana.

Recommended Prompts

1girl

Recommended Negative Prompts

(score_1, score_2, score_3), sketch, worst quality, low quality, deformed, censored, bad anatomy

Recommended Parameters

samplers

Euler a

steps

cfg

resolution

832x1216

other models

cfg_scale_boost (c071fe08ab71), 2DN-PLaY (5362fc4d25), CashMoney_v5 (cb5d75203d), bijutsubu_v1 (5a11c146a0), Illustrious-XL_SmoothFT_SPO (ac1faa25f6)

Recommended Hires (High-Resolution) Parameters

upscaler

8x_NMKD-Superscale_150000_G

upscale

1.5

steps

denoising strength

0.6

Tips

Experimentation is key to best use CFG Scale Boost LoRA, especially since it has varied effects on backgrounds and main subjects.

Strength values between 0.3 and 0.7 are recommended for this LoRA.

The realism LoRA can shift model style between 2D/anime and 3D/photorealistic using values from -1.0 (2D) to 1.0 (photo).

Use the realism LoRA to adjust originality or style of your model without overwriting base styles.

Value ranges for realism LoRA may need adjustment depending on your specific model to avoid over or under-effect.

Creator Sponsors

Join the Control LoRA Collection Discord for updates and community support

Announcing the release of the CFG Scale Boost LoRA! After much research I've discovered a method which allows you to pseudo-control the CFG scale of your gens using a special LoRA. It is listed under Pony but will work with every SDXL model (including Illustrious XL)! It has the biggest effect on the backgrounds, where it will fill them out and make the scenes feel more complete without frying your gens. Second, it has a detailer-like effect on the main subject, but not always! Experimentation is key. I recommend 0.3-0.7 strength. Released under EA as many sleepless nights were dedicated to this project, which I originally wasn't going to release publicly. Works with negative values as well, allowing you to control the composition and detail of your art to a fine degree, something not possible with typical detail LoRAs.

Control LoRA Collection

A collection of various control LoRAs for use in gens, merges and fine tunes.

Detail series

SDXL Enhance: controls the details and compositional elements of SDXL-base models.

CFG ScaleBoost: boost the CFG scale of your gens using a LORA without burning out your latents.

Realism series

Update: Finally added a Pony version!

Original description:

This LoRA will make your model either 2D/anime <-> 3D/photorealistic depending on how to use it.

For example <lora:realistic:-1.0> (negative values) will strengthen the 2D style and <lore:realistic:1.0> (positive values) will strengthen the photo style.

Potential uses:

Adjust the style of your model slightly to make it look more original
Make a realistic model more realistic (such as when the skin is too smooth/plastic)
Make a 2D/anime model more stylised by making it more 2D
Combined with the detail LoRA (linked in the resources) you can reduce details in your anime model and make it look more 2D to create a unique cell shaded style

This LoRA works well with blended models. It is neutral and won’t overwrite the style of your gens.

Tip: depending on what model you’re using will determine what range of values will be suitable. 1.0 might be too much for some models and not enough for others, for example. This is especially true of negative values.

Discord