Konosuba - この素晴らしい世界に爆焔を！ - Collection
Collection of character LoRA's for the kono subarashii sekai ni shufuku wo! (この素晴らしい世界に爆焔を！) series, intended to generate non-official / fan art style images.
Character LoRAs:
See versions for each character LoRA and recommended settings.
Multiple Characters LoRA:
Extra LoRA trained on images that included multiple characters, then merged with individual character LoRA's above.
This one works best with promps for multiple subjects, i.e. 2girls, or 1girl, 1boy.
Includes additional characters:
Kazuma Satou - 佐藤 和真
Wiz - ウィズ
Iris - アイリス
Eris - エリス
Shinjin Succubus - 新人サキュバス
Advanced tip (A1111 or Forge users)
I recommend trying the merge collection LoRA with the sd-webui-regional-prompter or sd-forge-couple extensions.
I used sd-forge-couple and adetailer with [SEP] tags for the group shot generations in the previews.
Great article by FallenIncursio on how that works:
How to Generate Multiple Different Characters, Mix Characters, and/or Minimize Color Contamination
Combined v2
This version combines all datasets, train together with additional revised tagging of quality and nsfw:
Quality modifiers:
masterpiece, best quality, low quality, aesthetic, very aesthetic
NSFW tags:
nsfw, explicit
Recommended prompt structure:
I recommend using around a ~0.8 weight.
Positive prompt:
{{tags}}
score_9, score_8_up, score_7_up, score_6_up, source anime, masterpiece, best quality, very aestheticNegative prompt:
score_4, score_5, score_6, monochrome, (censor), 3d, realistic, render, CGI, text, watermark, cartoon, sketch, greyscaleLicense
Description
Trained on AutismMix SDXL