P5 style anime illustration - bold outline, flat color, high contrast art style

YUKIZZ1YUKIZZ1
3
31
animemangapersona 5styleart styleillustrationloracomic stylesdxlp5
v1.0
model-image-0
model-image-1
model-image-2
model-image-3
model-image-4
model-image-5
model-image-6
model-image-7
model-image-8
model-image-9

Details

Download Files (1)

Model description

Description

This is a Persona 5 (P5) bold-line comic art style LoRA.

It is designed to replicate the iconic aesthetic of the original Persona 5: hard-edged bold black outlines, high-contrast hard shadows, and saturated colors.

It works with most mainstream anime/SDXL base models and can be used alongside other LoRAs.

Recommended weight: 0.6-1.0 (0.8 is tested as the optimal value; adjust within this range for varied results).

Usage Instructions

  1. Avoid conflicting prompts: Do not use prompts for realistic/soft-painting styles, as they will weaken the P5-style linework and color texture.

  2. Recommended prompts: Use with P5-style keywords and line/color enhancement terms for maximum accuracy.

    Suggested prompts:

    persona, sharp clean lineart, contour_deepening, bold outline, high_contrast

Recommended Settings

  • Base Model: Illustrious XL

  • Sampler: Euler a / Euler

  • Steps: 20-30 (recommended: 28)

  • LoRA Weight: 0.6-1.0

    • 0.9-1.0 for pure P5 style

    • 0.6-0.8 for mixed styles

  • CFG Scale: 3.5-7 (recommended: 3.5-5)

  • VAE: Use the base model’s default VAE if not specified.

Base Model Statement

This LoRA is trained on the open-source base model Illustrious XL (JANKU Trained Chenkin NoobAI derivative). All rights belong to the original base model author.

  • This LoRA does not modify or distribute any code or weights of the original base model, and is only a secondary training of a style LoRA based on it.

  • Use of this LoRA must comply with the original base model’s open-source license. Any violation of the original model’s copyright and distribution rules is prohibited.

  • This LoRA is not affiliated with the original base model author, nor does it represent the author’s views or endorsement.

Model Statement

  1. This LoRA is a personal non-commercial training work for learning and communication purposes only. Copyright of generated images is the sole responsibility of the user.

  2. All training materials used are either self-owned or legally licensed public materials, and do not infringe on any third-party rights.

  3. Commercial use of this LoRA or its derivatives (including merged, modified, or retrained versions) is strictly prohibited. This includes, but is not limited to, selling the model, hosting it on profit-making platforms, or paid distribution.

  4. Selling AI images generated with this LoRA directly is prohibited. If you wish to use generated works commercially, you must perform sufficient manual editing and secondary creation to make them independent original works. You are solely responsible for the commercial compliance of your works.

  5. Users must comply with local laws and regulations. Do not use this model to generate illegal, infringing, or harmful content. The author is not liable for any legal consequences arising from improper use.

Images made by this model