Sign In

Magic portrait

2.3k

Updated: Nov 1, 2024

stylephotofemaleportraitgirlswoman

Download

1 variant available

SafeTensor

166.89 MB

Verified:

Type
LoRA
Stats

252

2.1K

794

Reviews
Published

Sep 28, 2023

Base Model

SDXL 1.0

Training
Steps: 4,000
Epochs: 10
Hash
AutoV2
61AECB7B6A
default creator card background decoration
Followers - 38

38

Likes - 238

238

tmpo_0to158.png

SD1.5:
This LoCon model enhances posture and neutral emotion in large portraits.

It gives a more vivid expression to the subject's gaze. It also enhances the overall aesthetic of the image.

The model achieves a more realistic skin texture, resulting in a more lifelike appearance, while preserving and adding more fine detail to the image.

  • Recommended prompt: portrait, clouse-up, looking at viewer

  • Negative prompt: high contrast

  • Clip skip: 1 or 2

  • Resolution: 768x768