Doubled the training images, still not satisfied with the result, but seems to generate less deformations. Use with 0.7 strength on URPM or 1 on anime models.







This Lora is trained on 32 photos of women receiving ??? in their mouths. There are two "tags" that can recall the lora: "?????????????" and "drinking ???? from a man's cock"
The 50version is the most reliable since it doesn't destroy the images. The 75 and the 100 are to be used with 0.7 strenght or they will look like the original photos.
Doubled the training images, still not satisfied with the result, but seems to generate less deformations. Use with 0.7 strength on URPM or 1 on anime models.
1. Права на репостнутые модели принадлежат оригинальным создателям.
2. Для подтверждения авторства модели создателям следует обратиться к сотрудникам SeaArt AI через официальные каналы. Нажмите, чтобы подтвердить
