V2.1Upscaled some images and deleted the blurriest ones in set. The model now generates lamia cosplay correctly in most cases, and the overall image clarity has improved.V2Datasets are the same, but this version was trained on the WAI-illustrious-SDXL-17 checkpoint. Trigger words haven't changed.Note: The LoRA can sometimes produce blurry faces.V1It works, ¯\(ツ)/¯

Vexoria the sun eater (15 Outfits)

Files with SHA256: d54ae8a5c1ace7888cc64c96ea648a0ad2e99709df481125dcd1fb7edbdd05fb