Tensor.Art

Blackwork Style | LoCon

LOCON
Original


Updated:

blur image
(masterpiece, best quality), BlackworkStyleManityro, <lora:BlackworkStyle_V1-Manityro-AdamW:1.0>, greyscale, monochrome, 1girl, from side, profile, moon, starry sky
  • (masterpiece, best quality), BlackworkStyleManityro, <lora:BlackworkStyle_V1-Manityro-AdamW:1.0>, greyscale, monochrome, 1girl, from side, profile, moon, starry sky
  • BlackworkStyleManityro, <lora:BlackworkStyle_V1-Manityro-AdamW:1.0>, outdoors, night sky, racoon, trashcan, monochrome, greyscale, white background,
  • (masterpiece, best quality), BlackworkStyleManityro, <lora:BlackworkStyle_V1-Manityro-AdamW:1.0>, greyscale, monochrome, 1boy, knight armor, helmet, from side, moon, castle, holding sword
  • BlackworkStyleManityro, <lora:BlackworkStyle_V1-Manityro-AdamW:1.0>, monochrome, greyscale, lineart, 1girl, holding bouquet,
  • (masterpiece, best quality), BlackworkStyleManityro, <lora:BlackworkStyle_V1-Manityro-AdamW:1.0>, monster, cat, horror \(theme\), greyscale,
  • (masterpiece, best quality), BlackworkStyleManityro, <lora:BlackworkStyle_V1-Manityro-AdamW:1.0>, greyscale, simple simple background, 1girl, from side, profile,
  • (masterpiece, best quality), BlackworkStyleManityro, <lora:BlackworkStyle_V1-Manityro-AdamW:1.0>, monster, cat, horror \(theme\), greyscale,
  • BlackworkStyleManityro, <lora:BlackworkStyle_V1-Manityro-AdamW:1.0>, monochrome, greyscale, upper body, IncrsRanni, wavy hair, cracked skin, (extra arms, extra faces:1.2), doll, joints, doll joints, <lora:RanniV4LoRA:0.9>, hat, white dress, cloak, black background
  • BlackworkStyleManityro, <lora:BlackworkStyle_V1-Manityro-AdamW:1.0>, monochrome, greyscale, lineart, astronaut in space, star \(sky\), spacecraft

I have opened up a request form

Blackwork Style based on Tattoo artworks. Flexible weight. Training done on NAI.

Trigger: "BlackworkStyleManityro, monochrome, greyscale" always needed.

Suggestions/notes:

  • The outputs vary a lot between checkpoint with some being more detailed and others more simple.

  • Experiment with prompts and models for fun results, above samples can give a base idea but the style can do a lot more.

  • Works with other loras but needs weight to be adjusted.

My first attempt at a style lora that a friend requested, turned out surprisingly close to what we wanted and was fun to play around with. I hope you all enjoy.

Feedback and reviews are always appreciated.

Nerdy training numbers:

Trained on D8Dreambooth trainer

Optimizer: AdamW

Training resolution: 512

Unet LR: 1

Tnec LR: 1

Unet weight decay: 0.012

Tenc weight decay: 0.016

100 Epochs - 13800 Steps

Trained on 138 images.

9,593
Runs
26
Downloads
Uploaded
Base ModelSD 1.5
Trigger Words

<p>Release version - Fun and random but can achieve what you want</p>

  • Download the model file

  • Use without crediting

  • Share merges of this model

  • Use different permissions on merges

Commercial use: None

Comments

Related Posts