2023/6/24 update
Update dataset. Retrain the model using new craft.
2023/3/1 update
Reduce loss to 0.09. Adjust rank to 32 to balance file size and quality.
2023/2/17 update
Use offset noise to improve generating ability for lighter/darker images. Reduce model file size to ~1MB. The preview images are generated using latest version of YesMix.
Introduction
A LORA model trained with ~110 images of Signora from Genshin Impact. The 784mb VAEs (NAI, Orangemix, Anything, Counterfeit) are recommended. 0.6~0.8 weights should be enough.
Description
Reduce loss to 0.09. Adjust rank to 32 to balance file size and quality.
FAQ
Comments (14)
用的也是yesmix1.5的模型,但是跑出来的图质感就是和你的例图不一样是什么原因呢,还加了什么科技吗?谢谢!
没加任何科技,只不过用了hires fix
@zakp 终于搞定了,好像是eta的问题,改到0.67就好了,谢谢!另外大佬p站无题那几张图是用什么lora跑的呢
@yaoge Fashion Girl
是NovelAI?
@zakp hires fix是什么?插件吗?最近没玩跑图,课落下的太多了
@zakp 你简介里提到的0.6-0.8,是哪个的权重?
@AK2019 当然是lora的权重
@AK2019 hires fix是高分辨率修复
What does zakp mean when he say "Reduce loss to 0.09. Adjust rank to 32 to balance file size and quality.?
This is some information of the training parameter. I guess this is nothing to do with using this LoRA
Details
Files
signora-lora-v2.5-32dim-16ep.safetensors
Mirrors
signoraGenshinImpact_v25.safetensors
signoraGenshinImpact_v25.safetensors
signora-lora-v2.5-32dim-16ep.safetensors
signoraGenshinImpact_v25.safetensors
signoraGenshinImpact_v25.safetensors
CH-signoraGenshinImpact_v25.safetensors
signoraGenshinImpact_v25.safetensors
signoraGenshinImpact_v25.safetensors
动漫单眼面罩 signoraGenshinImpact_v25.safetensors
signora-lora-v2.5-32dim-16ep.safetensors
signora-lora-v2.5-32dim-16ep.safetensors
signoraGenshinImpact_v25.safetensors
signoraGenshinImpact_v25.safetensors
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.













