Sign In

Renji - Wind Breaker

0

Updated: Oct 21, 2026

characteranimeguymalemanwind breaker

Not recently loaded

Download

1 variant available

bf16 SafeTensor

Renji_krea2.safetensors

BF16, good balance • 223.88 MB

Verified:

Type
LoRA
Stats

28

0

Reviews
Published

Oct 21, 2026

Base Model
Hash
AutoV201496CF271
Trigger Words
rnji,a young man with dark slicked-back hair with a short top knot and shaved undercut sides, narrow dark eyes, thin sharp eyebrows and a sharp jawline,
00005-2274918168.png

This LoRA is designed to generate Renji from Wind Breaker.

Available for Krea2 and Anima.

Recommended LoRA Strength Krea2: 0.5 – 1.1 - I recommend starting with 0.5

Recommended LoRA Strength Anima: 0.8 – 1.0

Trigger words: rnji

Base Look: KREA2: rnji, a young man with dark slicked-back hair with a short top knot and shaved undercut sides, narrow dark eyes, thin sharp eyebrows and a sharp jawline,

ANIMA: rnji, 1boy, male focus, solo, black hair, slicked back, topknot, undercut, shaved sides, black eyes, narrow eyes, sharp eyebrows,

Default Outfit (Optional): KREA2: wearing a white and grey zip-up hooded jacket with teal trim, a teal high collar, the hood draped over his shoulders and two teal straps running down from the shoulders through two flap chest pockets,

Default Outfit (Optional): ANIMA: white jacket, gray jacket, hooded jacket, hood down, teal trim, high collar, zipper, teal straps, chest pockets,

Change of Outfit (Optional): KREA2: replace the default outfit sentence with your own, for example "wearing a black gakuran school uniform with a stand-up collar and gold buttons,"

Change of Outfit (Optional): ANIMA: replace the default outfit tags with your own, for example "black gakuran, school uniform, stand-up collar, gold buttons,"

Piece by piece (default outfit): KREA2

  • Jacket: a white and grey zip-up hooded jacket with teal trim

  • Collar and hood: a teal high collar and the hood draped over his shoulders

  • Straps and pockets: two teal straps running down from the shoulders through two flap chest pockets

  • Pants: light grey pants with a teal stripe down the side, or baggy black pants and black boots

  • Back: a teal abstract skeletal graphic printed across the back and a small flap pocket with two vertical black straps on the lower back

ANIMA:

  • Jacket: white jacket, gray jacket, hooded jacket, teal trim,

  • Collar and hood: high collar, zipper, hood down,

  • Straps and pockets: teal straps, chest pockets,

  • Pants: light gray pants, teal stripe, or black pants, baggy pants, black boots,

  • Back: from behind, skeleton print, teal print, black straps,

Tips:

KREA2: write your prompt as natural sentences, starting with the trigger word and the style phrase "a young man with dark slicked-back hair with a short top knot...", then the base character sentence, then "wearing ..." with his default outfit or your own. Quality tags and negative prompts are not needed.

ANIMA: use Danbooru-style tags with the trigger word first, then the base character tags and the outfit tags you want. Quality tags: "masterpiece, best quality". The LoRA was trained with mixed captions (tags plus a natural-language sentence), so you can add one short sentence after the tags. The dataset is anime screenshots tagged "anime screencap, flat color": keep those tags for the anime look and replace them for other styles (for example "monochrome, greyscale, manga, halftone" for black and white manga).

BOTH: if the teal trim, straps or hood of his jacket leak into another outfit, lower the strength and describe the new outfit in detail (on Anima you can also add "teal trim, hooded jacket" to the negative prompt); if his face or hair lose detail (the top knot, the shaved sides), raise it. His default look is a half-lidded smirk: strong expressions (laughing, crying, fear, shouting) override it, while soft expressions tend to keep his narrow eyes.

Please check the gallery images to see the exact prompts and settings I used to get an idea of how it behaves. I'd love to see your generations, so feel free to share your results in the comments!

This LoRA was trained using the following base models:

  • krea2_raw_bf16

  • anima-base-v1.0

Tested in Forge Neo UI.