For best outputs:

Sampler: DPM++ 2M Karras

Steps: 20, cfg 6 ~ 7, CLIP skip 2.

VAE: kl-f8-anime2_fp16.safetensors

Pos prompt: add the classic keywords for quality ("masterpiece, best quality, epic res" ect) then although optional, highly recommended the LoRa's "more_details" (weight 0.8) and "FaceBeauty_qinglong_V3" (weight 1.5), both included in the example runs.

Neg prompt: "bad-picture-chill-75v, verybadimagenegative_v1.3, easynegative, (fat), (nsfw:1.4), (skinny), (nipples), (pregnant:1.4), (cropped in:1.4), (multiple view), (missing hand:1.4), (fused highheels, fused feet:1.4), extra hand, extra foot, (missing finger, extra fingers, fused fingers:1.4), (monochrome, greyscale), (text, signature, watermark:1.4), (mutated, bad composition, worst quality, low quality:1.4)", being the "bad-picture-chill-75v", "verybadimagenegative_v1.3" and "easynegative" embeddings included in the example runs.