Recommendations:<ul><li>With some weights it is required to set the LoRA value below 0.7, in this case it is worth specifying additional prompt parameters, for example, red hair, brown/red eye, ponytail, but for most the optimal values are from 0.7 - 1.0 and this is not required.</li><li>Keywords: Milly\(milly)\ or Milly\(milly\) for v3.0R, 1girl, hair ornament, brown eyes/red eyes, red hair, ponytail, hair between eyes, medium breasts.</li><li>CFG Scale within 5.0 - 8.0.</li></ul>About versions:What's 'D'?Developer version<ul><li>Just an intermediate version of LoRA.</li><li>Version trained on anime images.</li><li>Used with anything v4.5 and NED(NeverEndingDream).</li><li>Keywords as in release version.</li></ul>What's new in version 2.0?<ul><li>Dataset changed.</li><li>In LoRA v2.0R, VAE was not used during training, and other weights were used, unlike v2.0RV.</li></ul>Version 1.0.<ul><li>52 images.</li><li>156 steps per epoch, 14 epochs.</li></ul>If you have any tips for training LoRA models, I will be glad to read.Post your results below! ༼ つ ◕_◕ ༽つ

Milly (ミリィ) "The Aristocrat’s Otherworldly Adventure: Serving Gods Who Go Too Far"

Woop! Woop! (づ｡◕‿‿◕｡)づ What's new?<ul><li>Made changes to the dataset.</li><li>No VAE during model training.</li></ul>

Files with SHA256: 2c9ac9fd6a81262c89544f849a401c2e18900ca323d5a3ecb2954206cdafc571