Artist Style LoRA for Klongi

Description:

This is an artist style LoRA of Klongi.

Training Settings:

• Optimizer: Prodigy / Cosine

• Base Model: Illustrious v1.1

• Final Model: WAI

• Trigger Word(s): kl0ngi

• Training Steps/Epochs: 2925/25 (Clip skip 2)

• Dataset Size: 78

• Recommended Weight: 0.9-1.1 (1.0)

Sampler Settings:

• Steps: 30

• CFG: 4.5

• Sampler: Euler_a

Style positive design trigger words:

Oil painting style: masterpiece, best quality, amazing quality, painterly, visible brushstrokes, traditional media style, oil painting effect, soft edges, rich texture, blurry

Sketch style: pencil sketch, rough lineart, graphite texture, visible pencil strokes, hand-drawn, unfinished sketch, monochrome base

Sketch style 2: grayscale, blurry, film grain, pencil sketch, rough lineart, visible pencil strokes, hand-drawn style, crosshatching

Character negative design trigger words:

Notes:

I trained this on both Klongi's sketch style and oil painting style. The oil painting has more influence on the dataset, but if you want to get the sketch style use the above style trigger wording. I tend to leave the quality and masterpiece tags, but feel free to experiment! If you're attempting to do a close-up or upper body portrait, you may need to increase the strength of the LORA to 1.15.

Suggested Resources:

https://civitai.com/models/1001945 - (Detail Slider IL)

https://civitai.com/models/827184?modelVersionId=1612720 - WAI Checkpoint

Please remember to follow/support the artist if possible.