You can achieve either colored or monochrome results (exclude any color tags and use monochrome, greyscale, hatching \(texture\) ).
Artist's resources:
https://www.pixiv.net/en/users/4122
If you found the LoRA interesting or somehow useful, consider supporting me by joining my Telegram channel. I post thoughts there and do streams of LoRA making.
Best shot update:
WotB
Recommended samplers:
DPM SDE++ ★ - 18 steps
DPM++ 2S a - 16-20 steps
DDPM - ~40 steps
Recommended upscale settings - here.
Recommended checkpoint:
NAI-XL vpred1.0 2d accelerated "Final" version
Won't get any better or worse.
ILLUSTRIOUS UPDATE
Well who would've thought that making a good style lora of an artist who has little colored images and has no consistent coloring techniques is nearly impossible. I've spend 5 days trying to achieve good result but ended up unsatisfied again. Well, at least it is something.
Illustrious version is really different from the pony one, and I don't know exactly why is that.
Again, it would be great if you share your generated results and maybe leave a feedback.
Description
NOOB version.
Fixed dataset and new training settings.
DISCLAIMER (applies only to NOOB-AI version): use my parameters to achieve best results. I didn't test the Lora on other checkpoints, so keep that in mind.
Trained on https://civitai.com/models/833294?modelVersionId=1190596.
Was tested on https://civitai.com/models/1201815/nai-xl-vpred10-2d-accelerated, both "Final" and "style2" versions. Better CFG for "Final" is between 2.1-2.5, for "Style2" is 2.5-2.75. Recommended settings:
20 steps Euler DY CFG++ + Align Your Step CFG 2.4
Hiresfix settings:
Use Fatal Anime upscaler if output is too bright or Foolhardy in other cases. Hires steps 12, denoise from 0.2 to 0.6, upscale by 1.5, sampling method Euler Dy.
Enhance prompt string (put it after all your main prompt):
masterpiece, best_quality, very_awa, absurdres, very_as2, hdr, High_dynamic_range, rich_details, clear_shadows_and_highlights, intense, enhanced_contrast, highly_detailed, amazing_quality, very_aesthetic
Use "detailed background" if necessary.
FAQ
Looks like we don't have an active mirror for this file right now.
CivArchive is a community-maintained index — we catalog mirrors that volunteers upload to HuggingFace, torrents, and other public hosts. Looks like no one has uploaded a copy of this file yet.
Some files do get recovered over time through contributions. If you're looking for this one, feel free to ask in Discord, or help preserve it if you have a copy.
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.