For best outputs: Sampler: DPM++ 2M Karras. Steps: 30, cfg 5 ~ 7, CLIP skip 1. VAE: none required. Pos prompt: the model is set to realism, with realistic keywords can become even more realistic. Optional: add the LoRa "detailed_notrigger" with a weight of 1 at the end of your pos (found in most of my example runs). Neg prompt: can be left blank / what you don't want to see. Checkpoint Trained.