Sign In

Light novel-style illustration (ライトノベル系イラスト)

2
30
2
2
Verified:
SafeTensor
Type
LoRA
Stats
30
2
Reviews
Published
May 9, 2025
Base Model
Pony
Training
Steps: 1,024
Epochs: 10
Usage Tips
Clip Skip: 1
Strength: 1
Trigger Words
light_novel_illustration
Hash
AutoV2
592EA0D445
default creator card background decoration
Bronze Pony Badge
Raxephion's Avatar
Raxephion

🌸 Anime Light Novel Style Illustration (ライトノベル系イラスト)


🖌️ Overview

Anime Light Novel Style Illustration is a stylized LoRA designed to emulate the lush, expressive aesthetic found in Japanese light novel cover art — adapted specifically for pony models. This version brings a painterly, vibrant, and organic feel to generations and is intended as the first step in a long-term stylistic evolution.

Whether you're aiming for moody scenes, heartfelt character shots, or cinematic slice-of-life vibes, this LoRA helps transform generations into polished, artful compositions reminiscent of professionally illustrated light novel pages.

🔧 This is Version 1 of the style — expect refinements and expansions as the style evolves in future releases.


✨ Style Characteristics

  • Soft painterly textures blended with clean anime detailing

  • Organic gradients, bloom, and ambient lighting

  • Emphasis on expressive color theory (warm tones, golden hour, reflections)

  • Storybook-style framing and composition

  • Adaptable from whimsical to emotional or dramatic moods

  • Compatible with pony and anthro pony models, with stylization shifting subtly based on model used


🛠️ Usage Settings

Trigger Word:
light_novel_illustration

Style Amplification Tip:
To maximize effect, surround your prompt with:

"in the style of light_novel_illustration ...prompt... in the style of light_novel_illustration"


LoRA Strength (Sliding Scale):

  • 0.1 – 0.9: Subtle style polish

  • 1.0 – 1.2: Balanced light novel feel

  • 1.3 – 1.5: Strong painterly enhancement

  • 1.6 – 1.9: Maximum stylization (use with care — may distort detail)

⚠️ Strengths above 1.0 may become intense or jarring depending on your base model — experiment to find your aesthetic sweet spot.


⚙️ Technical Details

Samplers:

  • Euler

  • Euler A

  • DPM++ 2M

  • DPM++ 2M SDE (highly recommended for painterly diffusion)

Scheduler:

  • Karras

Steps:

  • 15+ (More steps yield more painterly finesse and detail)

CFG (Classifier-Free Guidance):

  • Suggested range: 0.3 – 0.8

  • Other values might also work, though untested


✅ Model Compatibility

  • Primarily designed for anime pony models

  • Compatible with most pony and anthro realism models

  • Produces realistic but stylized generations on realism bases

  • Best used in high-res fix pipelines for cleaner painterly texture

📜 Usage Disclaimer:
This LoRA is strictly for personal, non-commercial use only.
You are not allowed to repost, sell, merge, or otherwise profit from this model in any form.
It is allowed on CivitAI for personal use only. Unauthorized redistribution or monetization is prohibited.