An artist lora based on <a target="_blank" rel="ugc" href="https://twitter.com/ZumiDraws">zumi</a><ul><li>Works best at 0.8-1.0 strength</li><li>Use clip skip 2</li><li><a rel="ugc" href="https://huggingface.co/embed/EasyNegative/blob/main/EasyNegative.safetensors">Easynegative embedding</a></li><li>model used <a target="_blank" rel="ugc" href="https://civarchive.com/models/5245/kotosmix">kotomix</a></li><li>model used <a target="_blank" rel="ugc" href="https://huggingface.co/WarriorMama777/OrangeMixs/blob/main/Models/AbyssOrangeMix2/AbyssOrangeMix2_hard.safetensors">orange</a></li></ul><h3>Training settings</h3>copied from a very helpful anon125images15 repeats2 batchsize4 epochs3.2e-4 learning rate5e-5 text_encoder3.2e-4 unet_lr768 res128 dim64 network_alpha

zumi (zumidraws style lora)

Just an overall upgrade from v1<ul><li>removed triggerword requirement</li></ul>

Files with SHA256: 5ef31aebd6a162dabd790568f83ff7fa07effcbaa508fbbe518f04c1509618a6