模型素材分辨率达到1280*1920，所以清晰度应该比较好。第一次制作Pony模型，系统自动打标，实测效果尚可。期待反馈，以更新版本。测试大模型推荐 <a rel="ugc" href="https://civarchive.com/models/376450/duchaiten-pony-xl-no-score">duchaitenPonyXLNo_ponyNoScoreV50</a>，它所生成的微表情真的很可爱。本模型只为测试及学习使用，不得用于商业用途。The resolution of the model material reaches 1280 * 1920, so the clarity should be relatively good.This is my first time creating a Pony model, and the tips have not been manually intervened. The actual test results are still acceptable. Looking forward to feedback to update the version.The recommended test model is <a rel="ugc" href="https://civarchive.com/models/376450/duchaiten-pony-xl-no-score">duchaitenPonyXLNo_ponyNoScoreV50</a>, which generates cute micro expressions.This model is only for testing and learning purposes and cannot be used for commercial purposes.

888_CNgirl_XL_v01

模型人物佩戴了护目镜，出图时，可以在提示词中添加 no glasses 来取消，但是，随着采样步数的增加，护目镜出现的概率会变高。采样方式只测试了 SamplerTCD + BetaSamplingScheduler + Automatic CFG，效果不错。修脸似乎不是必须的，因为修脸会改变人物的微表情，当然，您也可以自行测试最佳的出图方式。经验不足，如有任何缺陷，还请诸位谅解。The model character is wearing goggles. When creating the image, you can add 'no glasses' in the prompt to cancel it. However, as the number of sampling steps increases, the probability of goggles appearing will increase.The sampling method only tested SamplerTCD+BetaSamplingScheduler+Automatic CFG, and the effect was good.FaceDetailer doesn't seem necessary as it can alter the character's micro expressions. Of course, you can also test the best way to create the image yourself.Lack of experience, if there are any shortcomings, please forgive me.

Files with SHA256: ff2e3fc62babc1d103b851cf3eca6db122cf471df66f32d6531992bab9768c29