After hundreds of hours and 30 iterations of training and refining, this LoRA was built from a high quality dataset of 123 images, covering Carrot in 8 different outfits, including her Sulong form, complete with her red eyes, flowing white hair and extremely long tail. Every single image in the dataset was cleaned, retouched, and made as anime-accurate as possible.
✅ Eye variations: brown_eyes, maroon_eyes, pink_eyes, red_eyes, and dark_sclera
✅ Sulong mode included
✅ Consistent across full-body and close-up images
✅ Best results at cfg scale 5 weight 0.9
This is my most polished LoRA so far, and I hope it helps others create high-quality generations of Carrot across a variety of scenes and styles.
I don’t mean to brag, I just wanted to share how much love and effort went into this. Even one like or a kind comment would mean the world to me.
Thank you for trying it out!