Krea2This Lora can generate the hymen of a real person's unfolded vagina. As this is the first time training Lora, and there are few training images for real hymen, this Lora has fingers to pull open the vagina, and its stability is also relatively poor,There is no version difference between the two Loras, they are from the same batch, only two relatively stable ones were selected and sent up<span style="background-color:rgb(31, 32, 35);color:rgb(140, 143, 163);font-family:-apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, Helvetica, Arial, sans-serif, &quot;Apple Color Emoji&quot;, &quot;Segoe UI Emoji&quot;;font-size:14px">Trigger Words：hymen中文：此lora可生成真人的展开阴道的处女膜，因为是第一次训练lora，加上真人处女膜的训练图较少，此lora带有手指拉开阴道，稳定性也差一些,两个lora没有版本区别，是一炉出的只是选了两个相对稳定发上来了触发词：hymen，但中文描述也可以识别MOSSSimplified the training dataset, the effect is decent，Also, two relatively stable versions from a batch will be released，Disadvantages: Due to the limited number of training images and poor stabilityFor now, I'm just training this Lora. My computer graphics card is not very good, and I rented a graphics card to train these two Loras. If I have to train the hymen Lora in the future, I will use the images generated by my Lora as the training set<span style="background-color:rgb(31, 32, 35);color:rgb(140, 143, 163);font-family:-apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, Helvetica, Arial, sans-serif, &quot;Apple Color Emoji&quot;, &quot;Segoe UI Emoji&quot;;font-size:14px">Trigger Words：hymenMOSS的版本精简了训练图集，效果还行同样还是发出一批次中两个相对较稳定的版本缺点：因为训练图少，还是稳定性差暂时就训练这个lora我的电脑显卡不是很好，这俩lora是我租用的显卡训练的，如果之后还要训练处女膜的lora，我会用我的lora出的图来当训练集了触发词：hymen，但中文描述也可以识别

Real hymen

MOSSSimplified the training dataset, the effect is decent，Also, two relatively stable versions from a batch will be released，Disadvantages: Due to the limited number of training images and poor stabilityFor now, I'm just training this Lora. My computer graphics card is not very good, and I rented a graphics card to train these two Loras. If I have to train the hymen Lora in the future, I will use the images generated by my Lora as the training setMOSS的版本精简了训练图集，效果还行同样还是发出一批次中两个相对较稳定的版本缺点：因为训练图少，还是稳定性差暂时就训练这个lora我的电脑显卡不是很好，这俩lora是我租用的显卡训练的，如果之后还要训练处女膜的lora，我会用我的lora出的图来当训练集了

Files with SHA256: 6a121470801f8aca3c5c4f89d5e94bfe3f16c18940dd9794fed25284aca86d20