<h1 id="read-the-damn-description!!!-ba3z0uhhc">Read the damn description!!!</h1>I recommend using the LoRA model mainly for inpaint. Inpaint steps:<ol><li>Mask the face (with or without hair)</li><li>Increase the batch count or batch size for a larger selection of variations</li><li>Play around a bit with the denoising strength</li><li>Done</li></ol>Tip: You can also process a face twice through the inpaint procedure. Example: Original Image (DenStr 0.3) → Result Image (DenStr 0.3) → Completed imageSince the model is trained on a resolution of 1280x1280, I recommend using it only in resolutions between 960x960 - 1536x1536.

Files with SHA256: 43352fd76f539eba31358b00ef39832c6fb9be5573585043a29b2bab3a434fd7