🎞 LoRA is trained purely on anime screencaps<hr />🗝 Activation tags are in the right panel📝 Note: Only babylonia outfits are part of a dataset she was trained on. Grey jacket outfit is not a part of the dataset, so it's purely up to the checkpoint how it generates.📸 Always use ADetailer to get detailed eyes🛠 Technical notes:<ul><li>LoRA shoud be compatible with Illu models &amp; NoobAI. (Not tested.)</li><li>LoRA is compatible with various styles - Lower the weight &amp; remove "anime screencap" from the prompt, if you are getting too strong anime style</li><li>Hi. res fix used on all sample images:<ul><li>Upscaler: R-ESRGAN 4x+ Anime6B</li><li>Steps: 10-15</li><li>Denoising: 0.3</li></ul></li></ul><hr />📬Leave a like, comment, review but first of all, ENJOY ;)

Files with SHA256: 03a71db145428c2f54395ae76faa00d2bae57be40e41dae9b43d565f2e61935e