This model was trained on 994 images I generated using different SD models and curated over many months.

My goal for this LoRA is to create a tool that avoids crunchy textures, supports soft gradients, and improves the accuracy of anatomical compositions.