ユーザーLoRA
owk_bsh_yy style | Illustrious
- アップロード日時
- 2024/11/30 23:56
- 採用数
- 35.56k
- レビュー
- 大好評 (2)
- 権限について
- 画像生成と共有を許可
- ユーザーによるモデルのダウンロードを許可する
- 生成画像の商用利用を許可する
詳細説明
Please support the original creator.This model was not created to harm or otherwise inconvenience the original creator nor Civitai or its users. It can be removed upon request.Model based on the amazing works of @owk_bsh_yy on Twitter/Skeb. They use this style in their ongoing story. If you're a fan of slow-burn stories, and/or if you enjoy this style, please give them some love on their active platforms.Model trained on WAI-ShuffleNoob. You may use different models with this lora. Recommended settings (external gen-comfyui & civit generator):lora weight: 0.8-1.0sampler: euler ascheduler: sgm_uniform/normalsteps: 20-40CFG: 5.0-7.0hires fixAfter testing the model with WAI-ShuffleNoob, I find that I get pretty clear and consistent results, which sometimes suffer from underexposure. I automatically adjust my output in my workflow to account for this. Adding the 'manga' tag to the prompt seems to alleviate this issue. Prompting for multiple characters in landscape may fail to work without some regulation tags. Please let me know of any other issue. Read the story on twitter/skeb !