It's lora made with 80 imagesrecommend strength of 0.7-0.8Trigger word: momoFor a more accurate look of momo, use posted trigger wordsThis sample image is an image made with my model, and the results may be very differentThere are cases where only the upper body can be seen because it is learned from an unfinished image. If you don't want to see only the upper body, put the upper body in the negative promptSometimes, the proportion of the human body is misaligned because of learning with incomplete images

Files with SHA256: 9b0f889c7abbb5c83930f34ac120ec3f4665fe96e6dd3ed37f5ef00d9c345c95