This model is trained based on the original concept of the base model, with the aim of enhancing the effects of certain specific actions, compositions, and poses. In short, it is an entertainment-oriented model~
该模型基于底模原有概念上进行训练，目的是增强某些特定动作/构图/姿态的效果，总而言之，是一个娱乐性的模型~
TAG:see-through_silhouette,towel,nude
example:see-through_silhouette,who,holding_towel,nude
raining model: Animagine XL V3.1
Inference model: Animagine XL V3.1
Training resolution: 832*1216/1024*1024
Recommended output resolution: 1024*1280
Recommended model weight: 0.8~1
Discord : https://discord.com/invite/CFpyHceF52
Recommended dataset collection and processing tools: https://github.com/deepghs/waifuc
Possibly the most user-friendly dataset processing UI currently in use:https://github.com/Eugeoter/waifuset