特别鸣谢 一个友人:阿荣
在挑选素材的过程中起到了决定性的因素。感谢有他才会有后来的Nffa lora
NFFA ⬅❤Like➕Follow
权重:0.7~1.0
可以和dogmastyle一起使用。建议权重 dogma 0.3+nffaV1.0 0.7
Weight: 0.7~ 1.0
Can be used with dogmastyle. Recommended weight dogmaV6 0.3 + nffaV1.0 0.7
一个光影lora 配合Nffa效果非常棒建议权重0.5
A light and shadow lora with Nffa effect is very good. The recommended weight is 0.5
https://civarchive.com/models/1206413?modelVersionId=1358601 ⬅ Lora
搭配使用 dogmaV6 0.3+nffaV1.0 0.7+concept_flasheffect-ill-vpmini 0.2~0.5
Use with dogmaV6 0.3+nffaV1.0 0.7+concept_flasheffect-ill-vpmini 0.2~0.5
正反向提示词可以参考例图
For forward and reverse prompt words, please refer to the example picture
本人制作的所有lora禁止用于商业用途!!!
特此说明!!!
All loras I created are prohibited from being used for commercial purposes!!!
This is to clarify!!!
这是一个pixiv上很有名的AI画师的画风lora,他的名字是Nffa。喜欢这个画风的可以去关注一下。希望你们能够喜欢
This is the painting style of lora, a famous AI artist on pixiv, and his name is Nffa. If you like this painting style, you can pay attention. Hope you like it.
Description
FAQ
Comments (8)
他的风格是制作好原始图片后,再通过comfyui批量反推提示词然后生图覆盖上一层画风。感兴趣可以尝试一下
好的 我可以尝试下~~~~~
@kiratian 早期有人在C站发布过暗角的lora,但是现在不见了。盲猜是先通过nai生成原始图片再通过comfyui,反推提示词,图生图覆盖风格,所以打造出特有的风格,一般人还真模仿不了。即使是lora也不太行。
@hit 我最近也在尝试,能再多详细讲图生图覆盖风格的具体细节吗?是不是要对图片局部做调整而不是整张图片?
请问有配套的workflow分享吗,自己调的总是不太能复现
不好意思 我用的是webui
Has your cunnystyle been taken down?
Yes. After adjustment, I will put it back on the shelves.
