Shitty SDXL loras

Finetune-extracted styles (Illustrious v0.1)

Mosha style lora, 67 image dataset w/ 5x repeats
1 kept token: artmosha
Training .ps1 script

• Mosha v1 e12 64d 32c fixed 459mb
• Mosha v1 e12 64d 32c fixed 346mb Smaller version without TE thanks to anons bodged extract script
  • img1 img2 img3 img4
  • finetune extract vs cursed old lora bake w/ same dataset, not a fair comparison
  • token on/off

B-LoRA—inspired styles (Pony Diffusion V6 XL)

Deadnoodle style lora
Random anons dataset, didn't test this very much
Trained with score_9, source_anime

• Deadnoodle v1 e43 out0to5_in12356_te1te2
  • img1

Made in Abyss manga style lora
Trained with score_9, source_anime

• Made in Abyss v5 e10 out0to5_in12356_te1te2
  • img1 img2 img3 img4 img5 img6

B-LoRA—inspired styles workflow (Pony Diffusion V6 XL)

Just the basic workflow, I have no clue what dataset sizes, learning rates, algos etc. are actually the best.
This set of sliced traits is just what I ended up with, it may not be optimal for everything. Do some experiments!

Training (sd-scripts):

1. Copy this blora_conv_ffnet_outallbut0_inall_mid.toml file to use as a preset during training, you'll need lycoris.
2. Example .ps1 file using LoKr algo,Uses roughly 15-18gb VRAM, could be optimized more. You can increase the factor to 3 or 4 if you're desperate, I saw improvements going from 4 to 2 in early testing but I haven't tried again with final settings.
3. The .ps1 above uses immiscible noise, so you'll have to merge this pull request https://github.com/kohya-ss/sd-scripts/pull/1395:
   
   ⎗

   ✓

   1 2	git fetch origin pull/1395/head:immis_noise git merge immis_noise

Slicing:

1. Clone https://github.com/ThereforeGames/blora_for_kohya to your sd-scripts folder, the repo has more detailed install/usage instructions.
   1.1 Replace the slicer with this modified one blora_slicer.py if you want the lora to retain it's metadata.
2. Add this to your blora_traits.json
   
   ⎗

   ✓

   1 2 3 4 5

   "out012345_in12356_te1_te2": { "whitelist": ["lora_te1", "lora_te2", "lora_unet_output_blocks_0_1_", "lora_unet_output_blocks_1_1_", "lora_unet_output_blocks_2_1_", "lora_unet_output_blocks_3_1_", "lora_unet_output_blocks_4_1_", "lora_unet_output_blocks_5_1_", "lora_unet_input_blocks_1_1_", "lora_unet_input_blocks_2_1_", "lora_unet_input_blocks_3_1_", "lora_unet_input_blocks_5_1_", "lora_unet_input_blocks_6_1_"], "blacklist": ["proj_in", "proj_out", "alpha"] }

3. Run the slicer on your chosen epochs using the above --traits and you're done!

   python blora_slicer.py --loras ./input/{file} --traits out012345_in12356_te1_te2 --debug --output_path=./output/{no_safetensors}_out0to5_in12356_te1te2.safetensors

   Personally I add "input" and "output" folders to my blora folder and use this handy python script folders_blora_slicer.py to run the script on all the loras in the input folder. It's very useful when you're testing multiple presets and multiple epochs.
   

   ⎗

   ✓

   1 2 3 4 5 6

   in your sd-scripts venv: cd .\blora_for_kohya python .\folders_blora_slicer.py -> grab the results from the output folder

Styles (Pony Diffusion V6 XL)

Trained with score_9, source_anime

Mosha with new oven settings for baking, sketch is a good tag to use. V9 with shittier tagging was somehow superior at 1girl, standing but had some other issues with prompting. Hands often have issues in both.

• Mosha v11 e25
  • img1 img2 img3
• Alt. bake Mosha v9 e28
  • v9 - v11 img

Trained with score_9, source_anime, eu03

Quick bake with the 5th eu03 dataset from anon, mostly for science. e16 is probably the one you want.

• eu03 e16
• eu03 e28
  • grid1

Trained with score_9, source_anime, tsukushi akihito

9th epoch is probably the best for normal use. Last epoch colors/shading go a bit too hard but it's great for spooky stuff. 1.0 weight works fine but going down to 0.8 brings back more details to the backgrounds.

• Made in Abyss 9th epoch
  • img1 img2 grid1 grid2
• Made in Abyss last epoch
  • img1 img2 grid1 grid2
• autismDPO extract lora

Trained with score_9, source_anime, moshimoshibe

• Moshimoshibe (token: moshimoshibe)
  • example
  • example

Shitty loras

Styles (LoKr + Pivotal)

Requires webui 1.7.0 / dev branch
Feels more forgiving to bake while retaining prompt responsiveness & less spaghetti fingers with pivotal? Including the token in prompt seems beneficial but might be cope.

• Raita (token: raita)
  • Has some issues with random impact effects at 1 weight, cba fixing the tags and rebaking.
  • example

Styles (LoKr)

• Asanagi
  • example

Styles

• Youjo Senki manga
  • example
• Isutoshi
  • example
• Mosha
  • example
• Shokugeki no Soma manga
  • example
• Kimura Takahiro (Code Geass official art)
  • example

Styles (hires)

Styles baked at 1200+ resolution, using these will generate mustard dolphin gas

• ShindoL
  • example
• Made in Abyss manga
  • example
• Mizuryu Kei
  • example
• Usagi Nagomu
  • example
  • example
• Kubo Tite
  • example
  • example
  • example
  • example
  • example

Concepts

Instant loss 2koma mating press, 2e4 adam and 2e6 ada. Which one is better? It's all gacha bullshit that depends on model/prompt/loras.
2e4 feels gentler on style and 2e6 feels a bit more consistent. I uploaded both so you can share the pain of indecision. More notes in metadata, tldr: controlnet good.

• Instant loss 2koma mating press 2e4
  • example
• Instant loss 2koma mating press 2e6
  • example
• Instant loss 2koma mating press embed

Questionable

Kaiji