PixAI Membership
PixAI
PixAI
Generate
Sign in
Home
Generation
Models
Studio
NEW
Contest
Create with Mio.2
Toolbox
NEW
Guide

User LoRA

DiT.2
Style

Style Over CFG Ver.dit2

#Original

Create

  • AI Image Generator
  • AI Animation Generator
  • Toolbox
  • Thematic generators
  • Train LoRA
  • Mio.2 Agent
  • Studio

Discover

  • Trending Tags
  • Ranking
  • Model Market
  • Contest
  • News

About

  • Guide
  • PixAI Blog
  • Tsubaki.2
  • Meet Mio, Our Mascot
  • Content Rules

Pricing & Help

  • Membership
  • Buy Credits
  • Contact

Mobile App

  • App Store
  • Google Play
© 2026 PixAI
  • Terms of Service
  • Privacy Policy
  • Copyright Policy
  • Specified Commercial Transactions Law
Style Over CFG Ver.dit2 - AI Model cover image
Trained by
Avatar image of M I
avatar-frame

M I

Train Your Own LoRA
Uploaded at
Apr 14, 2026, 11:29 AM
Uses
235
Earned
15,804
Reviews
No reviews
Trigger Words

overcfg style for stronger lines and contrast

Permissions
  • Allow image generation & sharing
  • Allow users to download this model
  • Allow commercial use of generated images

Recommended LoRAs

‧₊ ♪˚⊹ epne stained glass ‧₊ ♪˚⊹ - AI Model cover image

User LoRA

‧₊ ♪˚⊹ epne stained glass ‧₊ ♪˚⊹
11
Osono (Kiki's delivery service) - AI Model cover image

User LoRA

Osono (Kiki's delivery service)
9
平木花子-厕所里的花子VS屈强退魔师(Hiraki Hanako-Toilet Hanako-san VS Strong Exorcist) - AI Model cover image

User LoRA

平木花子-厕所里的花子VS屈强退魔师(Hiraki Hanako-Toilet Hanako-san VS Strong Exorcist)
23
Female Trafalgar Law - AI Model cover image

User LoRA

Female Trafalgar Law
9

Description

This LoRA is designed with the concept of establishing “CFG Excess” as a visual style.

Its behavior changes clearly depending on the strength, allowing intentional control over color gain and Color Runaway.

Strength behavior:

• 1.2 — Triggers intentional Color Runaway

• 1.0 — Upper stability limit (barely stable)

• 0.7 — OverCFG style (high saturation & contrast boost)

• 0.4 — Color enhancement (stronger hue and saturation gain)

• 0.2 — Color correction (subtle and natural color boost)

The LoRA is trained with strong emphasis on color-direction features and minimal influence on shapes.

As a result, it produces a unique behavior where “colors can go wild while shapes remain stable,” even at high strengths.

CFG過剰(OverCFG)を“スタイル”として確立することをコンセプトに設計したLoRAです。

強度によって挙動が明確に変化するのが特徴で、以下のように段階的に作用します。

• 1.2:Color Runaway(意図的暴走)を発生させます

• 1.0:ギリギリ安定する上限ライン

• 0.7:OverCFGスタイル(彩度・コントラスト強化)

• 0.4:色強化(彩度と色相の伸びが増加)

• 0.2:色補正(自然な色ゲインとして働く)

形状方向の影響は弱く、色方向のゲインが強く学習されているため、 “色だけ暴れて形が壊れない”という独特の挙動を示します。

Comments