Lokr trained in the DAGASI's art style. Based on the previous lora, some parameters were changed to improve the generalization. Please refer to the lora I uploaded earlier.
820 newer images were used for training, and tags were manually proofread. The training images include male and female furries and a small amount of non-furries. These images include some 【限定】 images released by Dagasi. Unlike the uncensored images in the first two parts that Dagasi teacher will release, these images are bar-coded, so I manually removed the code with PS. Because the training set is large enough, this lora can generate various characters such as furries, humans, and pure beasts, and the cooperation effect with other loras should also be good, which needs to be tested.
Dagasi's painting style is cute, and one of its characteristics is that almost every character has a blush on their face, so if you don't like blush, you can add it to the negative tags, but the effect will not be significant. In addition, all the characters generated by this lokr are relatively young, even the mature characters, this is unavoidable.
This lora also contains several Dagasi characters, as follows:
perro-kun:perro-kun, fox boy, fox ears, fox tail, yellow eyes, two-tone fur, orange fur, brown fur, hair bobbles, brown hair,
yuki-chan:yuki-chan, leopard girl, leopard print, snow leopard, leopard ears, leopard tail, blue eyes, light blue hair, multicolored hair, multicolored fur, blue fur, stomach mark (If you want "heart-shaped pattern" on her belly),
momo-chan:momo-chan, fox girl, fox ears, fox tail, ears down, red eyes, two-tone fur, brown fur, pink hair,
kohaku-chan:kohaku-chan, tiger girl, tiger ears, tiger tail, tiger print, multicolored fur, tiger stripes, white fur, orange fur, black fur, green eyes, streaked hair, black hair, two-tone hair, orange hair, short hair,
buchi:buchi-chan, hyena girl, hyena ears, multicolored hair, pink hair, brown hair, multicolored fur, brown fur, blue eyes, hairclip, hair bobbles, futanari (if nsfw),
haruka:haruka-chan, rabbbit girl, rabbit ears, rabbit tail, brown hair, carrot hair ornament, red eyes, ears down, short hair, one side up, asymmetrical bangs, two-tone fur, brown fur,
Existing problems: I don't know the specific differences between Lokr and Lora, and which one is better. I hope you can help test it. In addition, the background effect needs to be improved.
Weight: 0.6 - 0.9 It is generally not recommended to use weights of 1 or more.
Support the Artist: https://fantia.jp/fanclubs/34875
Enjoy.
使用Dagasi的画风训练的lokr。在先前的lora基础上,改变了一些参数,泛化性更好了。请参考我之前上传的lora。
使用820张较新的图片进行训练,手工校对tag。训练图像包含男性和女性furry以及少量非furry。在这些图像中包含了一些Dagasi发布的【限定】图像,不像Dagasi老师会发布的前两部分的无码图像,这些图像是有条形打码的,于是我用PS手工去码了。因为训练集足够大,该lora可以生成furry,人类,纯兽等各种角色,和其他lora的配合效果应该也不错,有待测试。
Dagasi的画风偏可爱,其中一特点是几乎每个角色脸上都会有脸红,所以,如果你不喜欢常态的脸红,可以将该tag加入负面,但效果不显著。另外,由本lokr生成的角色全部都偏幼,即使是成熟的角色,这是无法避免的。
这个模型还标注了几个Dagasi的角色,如下:
perro-kun:perro-kun, fox boy, fox ears, fox tail, yellow eyes, two-tone fur, orange fur, brown fur, hair bobbles, brown hair,
yuki-chan:yuki-chan, leopard girl, leopard print, snow leopard, leopard ears, leopard tail, blue eyes, light blue hair, multicolored hair, multicolored fur, blue fur, stomach mark (如果你想要还原她肚子上的心形花纹),
momo-chan:momo-chan, fox girl, fox ears, fox tail, ears down, red eyes, gradient eyes, two-tone fur, brown fur, pink hair,
kohaku-chan:kohaku-chan, tiger girl, tiger ears, tiger tail, tiger print, multicolored fur, tiger stripes, white fur, orange fur, black fur, green eyes, streaked hair, black hair, two-tone hair, orange hair, short hair,
buchi:buchi-chan, hyena girl, hyena ears, multicolored hair, pink hair, brown hair, multicolored fur, brown fur, blue eyes, hairclip, hair bobbles, futanari (如果想要看牛牛),
haruka:haruka-chan, rabbbit girl, rabbit ears, rabbit tail, brown hair, carrot hair ornament, red eyes, ears down, short hair, one side up, asymmetrical bangs, two-tone fur, brown fur,
现存问题:不清楚lokr和lora的具体区别体现在哪里,哪个更加优秀,希望各位也能帮忙测一测。另外,背景的效果有待提高。
Weight: 0.6 - 0.9 一般不推荐使用1或以上的权重。
Description
About this version
The training album is as of 【ブイズ牧場産モーモーミルク】
This version of lokr is based on Noob-V-Pred-1.0. This version does not have to use Trigger Words. If you want to use it, use (dagasistyle:0.6~1.0).
This version is based on the LoRa version I uploaded earlier, with a slightly larger dataset and some modifications to the image tags. It seems to be better than the LoRa version in terms of generalization, and remains to be tested.
(All sample images were generated with Noob-V-Pred-1.0 using ComfyUI and using hires-fix.) (Note: In the sample images, the weight of the quality-enhanced lora is lower, at 0.35)
Future updates: Training will be carried out after Mr. Dagasi updates a certain amount. It seems that models such as pony7 are about to be launched, and I will also train based on them.
训练图集截止至【ブイズ牧場産モーモーミルク】。
该版本lokr基于Noob-V-Pred-1.0。该版本不必须使用Trigger Words,如果要用,则用(dagasistyle:0.6~1.0)。
该版本基于我先前上传的lora,略微增加了数据集,对图片的tag进行了一些修改。在泛化性方面似乎比lora版本优秀,有待测试。
(示例图像全都用Noob-V-Pred-1.0,使用ComfyUI生成,经过了高清修复。)(注意:示例图片中,提升质量的lora的重量都是较低的,为0.35)
今后的更新:会在Dagasi老师更新一定量后再进行训练。似乎pony7等模型即将上线,我到时也会基于它们训练。
FAQ
Looks like we don't have an active mirror for this file right now.
CivArchive is a community-maintained index — we catalog mirrors that volunteers upload to HuggingFace, torrents, and other public hosts. Looks like no one has uploaded a copy of this file yet.
Some files do get recovered over time through contributions. If you're looking for this one, feel free to ask in Discord, or help preserve it if you have a copy.
Details
Available On (2 platforms)
Same model published on other platforms. May have additional downloads or version variants.