Diffus
Get started
models/Control LoRA Collection - NAI vpred fix

Control LoRA Collection - NAI vpred fix

advokat
|
8/29/2025
|
11:27:15 PM
| Discussion
High-detail digital artwork of a cybernetic girl with blue hair and reflective robotic armor sitting on the ground with a sword behind her, shown from above.
Cyberpunk female cyborg with white hair and oni horns holding a katana against a vivid yellow background, featuring mechanical arms and futuristic armor.
A tranquil lake scene featuring a small wooden boat with a girl wearing a hat and overalls sitting peacefully, surrounded by pink and white clouds reflecting in calm water, with green trees and boats along the shore under a soft blue sky.

Recommended Prompts

masterpiece, very aesthetic, best quality, newest, absurdres, highres, very awa, 1girl, solo

Recommended Negative Prompts

lowres, worst quality, low quality, bad anatomy, bad hands, multiple views, abstract, signature, furry, anthro, 2koma, 4koma, comic, manga, text, watermark, username, artist signature, patreon logo, patreon username, twitter username, buttons, tunnel, loli, crowd

Recommended Parameters

samplers

DPM++ SDE CFG++

steps

14

cfg

1.2

resolution

832x1216

other models

NAI_vpred_fix (75b55891ca03), noobaiXLNAIXL_vPred10Version (ea349eeae8)

Recommended Hires (High-Resolution) Parameters

upscaler

R-ESRGAN 4x+ Anime6B

upscale

2

steps

12

denoising strength

0.2

Tips

If the LoRA effect is too strong, start with 0.5 LoRA strength to correct issues without altering original tonality significantly.

CFG ScaleBoost LoRA works with negative values to control composition and detail finely, unlike typical detail LoRAs.

Realism LoRA values: negative values strengthen 2D style, positive values strengthen photorealistic style; adjust based on model suitability.

Excited to release a new quality control LoRA for NoobAI XL v-pred!

It took a month to develop this LoRA and package it into an easy-to-use format.

NoobAI XL is an excellent model, but suffers from a variety of issues - wherein you have to use smoothing, quality and detail LoRAs or fine-tunes to get high quality results.

This LoRA significantly improves coherency of the output latents and prompt obedience using a variety of techniques while retaining all model knowledge.

If the effect is too strong, start with 0.5 LoRA strength, which will fix many issues without affecting the original tonality as much.

Effect:

For reference this is the target image:

Source: https://civitai.com/images/46921580

You can see the LoRA brings out the prompting intention more clearly, and has significant details. The stength:0 image is very far from the intended result, unless you try many different seeds.

Control LoRA Collection

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

Model Fix series

NAI v-pred fix: fixes issues with NoobAI XL v-pred models

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.

Original rainbow messageL

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.

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

Contributor

advokat avatar

advokat

Previous
Halftone Style - Flux - V1
Next
Abstract Art - v2.0
Use this model

Model Details

Model type

LORA

Base model

NoobAI

Model version

NAI vpred fix

Model hash

f65a890feb

Creator

advokat's avatar
advokat

Discussion

Please log in to leave a comment.

Images by Control LoRA Collection - NAI vpred fix

High-detail digital artwork of a cybernetic girl with blue hair and reflective robotic armor sitting on the ground with a sword behind her, shown from above.
Cyberpunk female cyborg with white hair and oni horns holding a katana against a vivid yellow background, featuring mechanical arms and futuristic armor.
A tranquil lake scene featuring a small wooden boat with a girl wearing a hat and overalls sitting peacefully, surrounded by pink and white clouds reflecting in calm water, with green trees and boats along the shore under a soft blue sky.

anime Images

Anime-style image of a girl with long rainbow hair sitting cross-legged on a mossy log, gently petting a large white wolf resting its head in her lap, surrounded by an ancient foggy forest with glowing mushrooms.
Anime girl standing barefoot on an empty subway platform at 3AM, wearing a sailor uniform and holding a red spider lily, with holographic glitch effects and flickering fluorescent lights.
Young woman with red backpack gazes at a colossal industrial ring structure over a vibrant golden field and rolling hills under a bright blue sky with fluffy clouds in detailed anime style.
Back view of a woman with long pink hair leaning over a metal railing wearing black overalls and black Converse sneakers against a night cityscape with street lamps and bokeh lighting.
Anime styled image of Tsunade from Naruto wearing a sleeveless shirt and pants, smiling and reaching out outdoors in a garden pathway.
A woman meditating peacefully at the river edge in a lush jungle surrounded by big green plants, vivid orange flowers, and colorful paradise birds flying above under soft ambient lighting.
A cyberpunk woman with a robotic arm carefully trims a bonsai tree in a traditional Japanese room with wooden floors and sliding doors, anime style.
A mature Aka-Oni female with red skin and black horns, blue hair partially covering her face, rubbing one eye while holding a horned stuffed toy, in soft, detailed lighting.
A goddess of spring with very long multicolored hair wearing a flowing white dress sits peacefully among lotus flowers and vegetation at sunrise, with birds flying against a clear blue sky.
A woman with long brown hair in a white blouse and black checkered skirt walks on a narrow wooden bridge high above clouds under a bright blue sky filled with soaring birds.

detailer Images

Medieval female warrior kneeling with sword, detailed green dress and armor, short silver hair with red feathers, background of a medieval village.

photorealistic Images

Close-up photorealistic profile portrait of a dark elf female with deep charcoal-black skin, glowing blind white eyes, messy white ponytail, and intricate cybernetic Victorian-style filigree port and glowing circuitry.
A dark, photorealistic forest scene at night with colorful glowing berries beside a small pond reflecting the moon. A hand-painted 2D chibi creature resembling a fox with a tribal mask sits next to the water.
Close-up of a photorealistic horror-themed woman with glowing red eyes and sharp teeth, black hair, red nail polish, and wearing a dark kimono.
A young woman wearing a vibrant rainbow-colored skirt and high heels stands divided between blooming flowers in daylight and a dark, cinematic night background.
A small chibi kodama wood spirit releasing a glowing Chinese lantern into the night sky over a photorealistic pond and meadow with many floating lanterns and bokeh lights.
A figure cloaked in shadow stands before a towering wooden gate engulfed in bright flames in an ancient, misty forest, surrounded by burning embers and charred trees.
Photorealistic humanoid robot completely covered in overlapping PG-13 sticky notes with only the metallic face exposed in a dark blue-lit server room.
Photorealistic portrait of a cat girl with long purple hair and animal ears lying on a bed with white sheets in a Valentine's bedroom setting.
A woman meditating peacefully at the river edge in a lush jungle surrounded by big green plants, vivid orange flowers, and colorful paradise birds flying above under soft ambient lighting.
A highly detailed 3D geometric structure with intricate patterned lines glowing in orange, rendered with cinematic lighting and photorealistic effects.

tool Images

A furry humanoid character with a backpack and axe walking through a post-apocalyptic landscape. This is an AI generated image using Stable Diffusion.