real-life

For best outputs:

• Version 4.0:

  • Sampler: DPM++ 2M Karras.

  • Steps: 40, cfg 5, CLIP skip 2.

  • VAE: none required.

  • Pos prompt: nothing remarkable, just what you want to see. Model listens real well and doesn't need of too much weighting in most cases.

  • Neg prompt: can be left blank, or just what you don't want to see.

• Version 3.0:

  • Sampler: DPM++ 2M Karras.

  • Steps: 30, cfg 5, CLIP skip 2.

  • VAE: clearvae_v2.3_fp16.safetensors.

  • Pos prompt: add the LoRa's "more_details" with a weight of 0.8 and "FaceBeauty_qinglong_V3" with a weight of 1 at the end of your prompt (found in my v3.0 example runs).

  • Neg prompt: "(deformed, distorted, disfigured:1.3), poorly drawn, bad anatomy, wrong anatomy, extra limb, missing limb, floating limbs, (mutated hands and fingers:1.4), disconnected limbs, mutation, mutated, ugly, disgusting, blurry, amputation".

• What changes from v3.0 to v4.0: input_image feature, overall quality, model listening and realism improved.

Checkpoint Merge.