For best outputs: Sampler: DPM++ 2M Karras / DPM++ SDE Karras Steps: 30, cfg 7, CLIP skip 2. VAE: none required. Pos prompt: optional, add the LoRa's "more_details" with a weight of 0.7 and "FaceBeauty_qinglong_V3" with a weight of 1 at the end of your pos for some more of detail and clearer faces specially when at a distance. Neg prompt: make sure to add the embeddings "negative_hand-neg", "ng_deepnegative_v1_75t" and EasyNegativeV2" in your neg (included in my example runs for both versions). What changes from v3.0 to v4.0: input_image feature. Checkpoint Merge.