For best outputs: Sampler: DPM++ 3M SDE Karras. Steps: 55, cfg9.5, CLIP skip 1. VAE: sdxl_vae.safetensors Pos prompt: include the LoRa "add-detail-xl" with a weight of 1 at the end of your pos (included in my example runs). Neg prompt: "loli, nsfw, low quality, low detail, worst quality, face asymmetry, eyes asymmetry, deformed eyes, open mouth, text, EasyNegativeV2, unaestheticXL_hk1", being the "EasyNegativeV2" and "unaestheticXL_hk1" embeddings included in my example runs. This is the big brother of the SD1.5 Experience model, but this one is focused on different art styles instead of realism, included anime.