N°46 Release, 胡桃 - HuTao from Genshin Impact - 原神.
Put everything in Filter for seeing all images!
Click on Show more for Weight & Trigger recommendation, Also click on About This Version at Right Corner for more Info.
Character's Request & One of the biggest datasets like the other girls from Genshin impact that i did to this Day, All Manually at Max Quality Possible, Very good Quality Output.
About this Lora:
This Lora is trained on 574+ Images at the Max Quality Possible, (Orginal Art 4K + Official Art 4K), I sorted everything manually, Only picked the best possible. (SFW+)
NAI Used, Work with Majority of the Anime Model's.
Trained at 768x Resolution for Maximum Detail & Quality.
Flexibility is Not Bad/Good.
Good Precision on Official Default Skin & Can do Also Other Clothes that don't exist (Like HuTao Office Lady, Swimsuit).
Can do Original Art Easily.
Work with other Lora like Add Detail & Other. (See Examples images)
Probably can do in Realistic Cosplay (Not 100% Sure).
About Trigger Words & Prompts: HuTaoV4 (For Trigger) & for prompts just refer to examples image.
Weight recommended: 0.7-1. - Work perfectly at Weight 1, Best result: 0.8-1
Negative Embedding used: EasyNegative, BadPrompt2 and sometime negative_hand-neg.
Leave a Review & Share you're image in comments if you appreciate this Work :).
Description
Recommendation for Official Default Skin: I recommend to put: ((Very long hair, Multicolored Prompts)), For a Better Prescision.
For Official Default Skin: Official Skin = 0.8-1.0
For Flexibility: for like clothes that don't exist put 0.6-1 Weight, See my Examples images for this.
Also if you don't want something to appear, put it in Negative Prompt
FAQ
Comments (6)
Excellent Lora, amazing outputs. For this Lora, I can suggest HassakuHentai model over the others. Still works amazing with most of the models tho.
Nice 👍 Actually i retrained this Lora 3x time ahah, because i was not satisfied with the result before, but here it's the Most Quality & Prescision i got.
Only things that's miss is the specific Eyes symbol-shaped in Flower (Like all Lora's), For this it's obligatory to train a specific Lora's for this only parts, like i did for Daki Eyes ahah.
@KXRO I think her Lora is hard because of all the clothes, she has so many different things. When you render less clothes, output becomes better, which that's no problem for me lol.
@kaiai Ahah yea, that's the limitation of SD Model's 1.5 & Lora's at his Maximum possibility, It's become RNG At this time, Tho when SDXL Will be optimized for -8gb Vram Cards & Better SDXL Model's are optimized, this limitation will decrease and output will surpass x4 SD 1.5 Lora's, Quality & Composition will be Far Better with a Little Less Error.
@KXRO Hey, I just created my first LoRA ever (ofc after unsuccessful tries). You can check my profile if you interested :3 Don't ever forget to check the last image I created
Another amazing contribution! I got your content on alert. Keep it up!