I do not mean to be that person, but I feel this model is a bit under-tuned. Many key details are missing or end up as indistinct blobs -especially things like Zoe’s safety pins on her hat or the gold Grizzbolt pin bulk. Illustrious is definitely capable of reproducing small text and details when the LoRA is well-trained, so if these elements are not showing up, it likely means the dataset or training could use more attention. Just some honest feedback! I am looking forward to see an improved version in the future.