---
license: creativeml-openrail-m
language:
- en
library_name: diffusers
pipeline_tag: text-to-image
tags:
- Lora
- stable-diffusion
---
---
- ## Introduction of LoraByTanger - 介绍
欢迎使用由Tanger训练的Lora模型,此模型库包括了本人当前训练并筛选出的所有人物Lora模型,目前以原神角色为主,但未来也会包括更多的游戏人物与二次元角色(一切以个人喜好与XP为准)。
Welcome to LoraByTanger. This model library includes all the Lora models I have trained and selected so far. Currently, the GenShin Impact characters are main, but in the future, more game characters and anime characters will be included (all subject to personal preference and XP).
- ## Composition - 组成
每个Lora文件夹包含了-Each Lora folder contains:
1. Lora Model
2. test.png (Generated by ["AbyssOrangeMix2_hard.safetensors"](https://huggingface.co/WarriorMama777/OrangeMixs/tree/main/Models/AbyssOrangeMix2))
3. Good pics * n (Generated by other models)
4. "xx配置文件.json" means the training parameters used in this Lora.
- ## Suggestions - 建议
适当降低权重使用可确保一定的泛化能力(换衣服),如果出现某个原配服饰无法消除,可在负面词会添加对应tag~
适当调高cfg有益于提升画面精细程度,且不同模型有不同的适宜区间。
Appropriately reduce the use of weight can ensure a certain generalization ability (change clothes).
If there is a certain original clothing can not be eliminated, please add the corresponding tag to the negative promote~
Adjust the cfg is beneficial to improve the picture quality, and different models have different appropriate interval.
- ## Examples - 例图
所有测试图均由["AbyssOrangeMix2_hard"](https://huggingface.co/WarriorMama777/OrangeMixs/tree/main/Models/AbyssOrangeMix2)生成,至于为什么全是死库水,可能(这就是爱)我只是想测试一下Lora的泛化能力。如果想要完全还原角色,建议随便找一张角色原图反推tag后,加入prompt以还原角色,或者干脆不额外使用服装等影响角色形象的tag。
(可在文件列表中找到原图,并放入WebUi查看关键词等信息) - (You can find the original image in the file list, and put WebUi to view keywords and other information)
- # Candace_坎蒂丝
- # Kamisato_Ayaka_神里凌华
- # Layla_莱依拉
- # Yanfei_烟绯
- # Ranni_菈妮
- # Yoimiya_霄宫
- # Ningguang_凝光
- # Noelle_诺艾尔
- # Nilou_妮露
- # Mobius_梅比乌斯
又又又重炼几炉。。。基本实现了三套服装稳定合并到一个,但具体还原和使用效果可能还需要模型配合。(Basically, the three sets of clothing have been stably combined into one, but the specific restoration and use effect may need the cooperation of the model.)
Tags:
- 海洋(Sea):(twintails, mbsw, blue long skirt:1.1),(blue dress:1.1), horns,(green hair:1.3), detailed eyes,
- 沙漠(Desert):(mbac, arabian clothes, white cat ears hood:1.2),egyptian clothes, white-egyptian-dress, (white dress:1.3), desert,(mouth veil,veil:1.3), bandages, (green hair:1.3),gold jewelry,detailed eyes,
- 原皮(Oringin):(mbor, black dress :1.1),(black Trident Crown:1.2),black headpiece,green hair,detailed eyes,
四个版本的主要区别在于底模,以及是否使用正则化。均在模型名称中有体现(The main differences between the four versions are the base mold and whether regularization is used. are reflected in the model name):
- -any4.5:Trained by anything4.5
- -nai:Trained by animefull-latest
- -reg:Regularization was used
- # (v3)-any4.5-mobius(mbor_mbac_mbsw_mbothers)
- # (v4)-nai-mobius(mbor_mbac_mbsw_mbothers)-000016
- # (v5)-nai-reg-mobius(mbor_mbac_mbsw_mbothers)-000010
- # (v6)-any4.5-reg-mobius(mbor_mbac_mbsw_mbothers)
- # Shenhe-申鹤
根据训练材料和训练程度分了四个版本,大家酌情使用~
推荐tag:shenhe,long silver hair, hair ornament,medium breasts, blue eyes, braid, earrings, gloves, (hair over one eye:1.2), jewelry, looking at viewer, navel, solo,
more