Aesthetic detailed

A LoRA that aims to enhance the light and background of images

Base model: Chenkin noob xl v0.5

Trigger: aesthetic detailed

Latest release:

v0.1_large:

Trained on larger LoRA network(dim = 96)

Preform slightly better in overall image composition.

Versions:

v0.1_large_preview:

resolution: 1152x1152

dataset: 1983 images

training time: 6 GPU hours on RTX5090D

suggest to use 1.0 or lower strength.

v0.1_preview:

resolution: 1216x1216,1408x892

dataset: 1983 images

training time: 5 GPU hours on RTX5090D

suggest to use 1.0 or lower strength.

Note:

watermark and copyright problem will be fixed in V1.

The LoRA is in preview version, further training in progress, welcome to give suggestions.