PixAI Membership

📢 Login Issue Resolved — Earlier today, some users may have unexpectedly been logged out of their PixAI accounts, and this was due to a known ser…

View details

User LoRA

XL
Style

Artist style-🐦(v2)

Artist style-🐦(v2) - AI Model cover image
Trained by
Avatar image of Coore

Coore

Train Your Own LoRA
Uploaded at
Mar 1, 2026, 11:40 AM
Uses
54
Earned
5,725
Reviews
Excellent  (1)
Trigger Words

mq_style,

Permissions
  • Allow image generation & sharing
  • Allow users to download this model
  • Allow commercial use of generated images

Recommended LoRAs

Chiester 45 | Umineko When They Cry - AI Model cover image

User LoRA

Chiester 45 | Umineko When They Cry
8
Fox Mccloud (starfox) - AI Model cover image

User LoRA

Fox Mccloud (starfox)
13
バイオハザード9 エミリー / Resident Evil 9 Emily - AI Model cover image

User LoRA

バイオハザード9 エミリー / Resident Evil 9 Emily
4
Seres / シアリーズ (Tales of Berseria) [IllustriousXL] - AI Model cover image

User LoRA

Seres / シアリーズ (Tales of Berseria) [IllustriousXL]
4

Description

增强版，手动处理数据集，不一定要去除halo标签，但是还是添加比较好，因为PixAI的系统好像把Blue_Archieve的标签打上去了，只要出图的时候不使用它并且将halo加入反提示词就没多大问题。 建议采样器：Eular A（颜色较淡） DPM++ 2M（颜色较深） 建议权重：0.7-1.1，最合适就是0.85 建议步数：28-36，最合适是32 建议CFG：8 提示词除了质量提示词以外，最好不超过35个，并且提示词尽量不要太长，将长提示词分割成更小的提示词效果更好。 PS：最好加上 petite 并给它1.2到1.5的高权重，这样的出图质量最高，因为原数据集的所有图片都有 petite 这个标签

Comments

Artwork made by Artist style-🐦(v2)

HomeModelsGenerateMio.2Profile