我们需要真正理解女性年龄的模型而不是只能生成萝莉！
We need a model that truly understands female age, not one that only generates "loli"!
所以这是一个旨在生成成熟女性并且试图理解多样的女性年龄，而不是只能生成小孩、女孩、萝莉的模型。
Therefore, this is a model designed to generate mature female and attempts to understand the diverse ages of women, rather than only generating "children, girls, or loli."
我的推特:@ekll01
My Twitter: @ekll01
版本说明请查看About this version
Please see "About this version" for the version description.
Please refer to the workflow that I have disclosed.
目前支持的年龄tag
女性：lxlx、mesugaki、bishoujo、lady、mature lady
男性：shxtx、bishounen、gentleman
请您用以上tag代替1girl、1boy。
Currently Supported Age Tags
For females: lxlx, mesugaki, bishoujo, lady, mature lady
For males: shxtx, bishounen, gentleman
Please use these tags instead of 1girl or 1boy.
Recommended Settings（general）
CFG: 5
Steps: 40
Sampling: Euler a
Scheduler: simple
Resolution: 832*1216 above
希望喜欢这个模型的朋友能够量力为模型进行一下Bid投资。这不是请求，但是如果有人能够帮助模型进行Bid，我会十分感谢！
I hope that friends who like this model can consider making a "Bid" investment in the model according to their ability. This is not a request, but I would be very grateful if someone could help the model with a "Bid."
Description
基于1.2.1的略微改动版本，并且增加oil元素。
Based on the slightly modified version 1.2.1, add the oil element.
FAQ
Comments (48)
Paywall? Really? Sorry, not supporting that.
I have set it to the minimum value. Theoretically, only need to sign in for 4 days to accumulate 100buzz. This is to cover the training expenses, not for my own profit. Please understand. It will automatically become public in three days.
You can’t always expect to get anything for free
@EKLL It's not about sum, it's about principle. Paywalls are a big factor of ruining the Internet, and I can't support that. I'm sure there are people who don't care and will give buzz, so don't worry.
@Rienie I'd rather donate 2x more than use the paywall - but the author didn't ask for donations and set a paywall instead.
@dobomex761604 This is not a payment; it's merely an EA...
@dobomex761604 Donation is way of paying, because I can develop a special version for donors instead of providing it to non-donors.
Moreover, I can't get any cash from buzz. All buzz is returned to civitai's cloud computing.
@dobomex761604 The unfortunate reality is that since Civitai introduced the auction feature, I've hardly received any Buzz donations. In the 18 day, I only got 339 Buzz, which doesn't even cover the expense of one basic LoRA training session. I'm not desperate to enable Early Access just for financial gain (especially since Buzz can't be converted to cash), but I feel like I'm left with no other choice.
@EKLL I did my choice, and I won't change it. I hope it helps you, but my opinion of paywalling (early access, whatever it's called) won't change.
I don't understand why people feel ashamed of explaining that they need donations and asking for them in model descriptions, for example. Paywall is not the way to do it.
@dobomex761604 You need to take into account that in some countries, donation channels are very limited, or in some countries, there are legal restrictions, which also include taxes and so on
EKLL Ignore this bozo. Great work on the model, it's essential to character variety in my stories. I will do my part with some buzz for you!
I have seen a lot of comments about your model. You don't need to pay too much attention to the performance of the model. Follow your original idea and make the model you want to use yourself. The greatest value of your model is to provide a common gathering place for us enthusiasts in an environment full of loli models. There is no doubt that your model has an audience.Don't worry about whether there are new models for technological and expressive breakthroughs, you can extend different style models on the basis of existing stable models. Just update according to your own preferences, and don't have to be coerced by others' strict requirements.
Your model has always been the best in my eyes!
Thank you very much.
登录领取的buzz无法使用……我先去绑个卡，不知道支不支持银联。
谢谢，但是不用麻烦了，已经凑够训练一次的buzz了。
关于这次的新版本开启ea的说明：
本次开启ea的原因是为了筹集civitai云计算训练lora所需的buzz。以260张图片为例，一次训练所需的buzz大概为4000。当然260张图片作为数据集来说很少，而且一次训练也是不够的。
选择100buzz是因为100buzz是能选择的最小的单位数值，如果可能我希望可以选择25buzz。
目前已经筹集到了9000buzz。
本次筹集的buzz将用来训练新版本的GrownRuler，距离上一次更新已经两个月了。
此外还有1.2中使用的新训练。
正如1.2发布时说的，在“去除loli”相关的训练有新的突破前，我将会暂时停止Mature ritual的模型制作。除非像是1.2.1那样为了解决bug而更新的修复版本。
Regarding the Early Access (EA) for this new version:
The reason for enabling EA this time is to raise the required Buzz for training LoRA models on Civitai's cloud computing platform. For instance, a single training session with 260 images requires approximately 4,000 Buzz. Of course, a dataset of 260 images is quite small, and a single training session is insufficient.
The choice of 100 Buzz is because it is the smallest unit that can be selected. If it were possible, I would have preferred to set it at 25 Buzz.
Currently, 9,000 Buzz has been raised.
The Buzz collected from this EA will be used to train the new version of GrownRuler, as it has been two months since the last update.
Additionally, there will be new training methods used in version 1.2.
As mentioned in the 1.2 release, I will temporarily suspend the creation of the Mature Ritual model until there are new breakthroughs in the training related to "loli removal." The only exception would be for maintenance releases to fix bugs, such as version 1.2.1.
Yes, with this Checkpoint, you no longer need to look for the damn mature lora
Thank you
大佬...那个...我能不能问一个问题，就是，我之前收藏了您的这个模型，也点赞了，但是按名称搜索您的MOD搜不到，收藏里也不见了，但仍然可以通过点击您其他的模型和LORA或者点击您的个人主页顺藤摸瓜进入此模型的页面，其他一部分作者的模型也是这种情况，这是什么原因呢？
1.被“审查”了，例图基本都被标记为R级了，虽然我并不觉得选取的例图能被算作R级。。。您可以看一下账户设置是不是允许查看R级内容。
2.热度太低又没有参与推广竞拍被降权了，这个我也没办法。。。
我这边是没改动过对于这个模型的设置。。。
EKLL 主要是我一直没改动过任何设置，突然就这样了。在您主页model界面里，此MOD也消失了。同样的，其他同样情况的作者主页也是这样，好像有点被“屏蔽”掉的意味，如果主页没有之前示例图的话，是很难找到被“屏蔽”的模型页面的，突然被这种方式处理，感觉很冤枉。
simonfranck7729711 我也很难受，但也没什么解决方案，可能挂个推广会好一点？我不确定，现在也没什么其他地方可以发布
EKLL 话说有没有更换示例图可以重新申诉之类的操作渠道呢？这么好的模型现在无法被人看到，很令人惋惜。
simonfranck7729711 例图倒是可以更换，但我也不知道啥样的能不被标R级
EKLL 对不起...是我的问题，不知道怎么回事，我设置被复原了，我说咋看不到部分模型了呢，原来它又把R级给关了...，话说大佬，我想顺便问一下，1.21是不是和1.2一样，也不需要添加质量提示词呢？只有11b和以前的需要对吗？
simonfranck7729711 对，任何质量相关的提示词都是没必要的
Thank you so much！I don't need use other milf style lora, I have this！ (o゜▽゜)o☆
Thank you
is there a way to make the faces less slender? i've been looking for a lora that makes women older but the faces on this one are too narrow.
This is not a lora... But you can try "slender face, round face, oval face, baby face, inverted triangle face, heart-shaped face, diamond face" square face, long face, triangular face,”
他的意思是说，人物的身体足够成熟，但是面部曲线，轮廓还是太过年轻，まめほしお https://www.pixiv.net/users/1619152。まめほしお，我给您推荐一下这位的画风，他的面容，描写的十分好，希望能给您一些启发，或者做一下这个风格的lora。
@1124992235888 这是底层错误了，特定tag就是会无视年龄强制萝莉化
肢体还是不太协调，对提示词的反应也不太敏感
肢体不协调是指什么，提示词在1.3会改进
EKLL grabbing another's breasts的时候经常会出现女的摸自己的胸。还有missionary体位的时候，提示词里有hug，人物的手臂会连在一起。
ddmf098 如果提示词有solo的情况是不行的，因为another's包含了两个人的概念。而solo否决了这个概念
EKLL 提示词里没有solo，但不管怎么样，这个已经是很好的熟女模型了
ddmf098 我测试grabbing another's breasts在没有solo时没问题。。。您可以提供一下完整的提示词我试试
EKLL uncensored,masterpiece,best quality,very aesthetic,absurdres,newest,amazing quality,highres,solo focus,from side,
standing,standing sex,sex from behind,bent over,grabbing another's breast,
mature female,detailed face,detailed eyes,(long face,slender face),brown eyes,orange hair,female pubic hair,nipples,(huge breasts),
full-face blush,moaning,heavy breathing,drooling,saliva,embarrassed,shy,looking back,
BREAK
1boy,(shota),eugeo,(faceless male:1.2),huge penis,testicles,younger man,young boy,younger boy,young man,
large penis,
love hotel,hotel room,night,against wall,
trembling,spasm,motion lines,cumdrip,
prefect lighting,soft lighting,cinematic lighting,
depth of field,blurry background,
masterpiece,best quality,very aesthetic,absurdres,newest,amazing quality,highres, 感觉男女有体型差的时候，肢体会很不协调
ddmf098 先放弃所有的“masterpiece,best quality,very aesthetic,absurdres,newest,amazing quality,highres”之类的词
ddmf098 尤其是“masterpiece,best quality”这两个词千万不要用
EKLL 我个人感觉好像用不用区别不大？不管怎么样还是感谢提供这么好的熟女模型
ddmf098 有些条件下会让角色年龄变小
probably my best checkpoint experience to use. the result is exactly what I want. If I use the exact danbooru tag, this model follows it to a T even if not using lora. also the Lora understanding is top notch, it let the Lora take the stage and not interfere it too much with its base style. It also accept pony lora easily and well to my surprise. Best model I've used, thank you for this.
Thank you for your use! Have a good time
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.