LoRA for Asian girls in the 1990s, based on SDXL1.0 training.

Trigger word: cn1990s

Training materials: 36 old photos from the 1990s, after simple processing, the borders and texts were removed, and the pixels were adjusted.

Base model: sd_xl_base_1.0_0.9vae.safetensors

The generalization performance is average, and it is recommended to set the strength to 0.4~0.6. If the generated pixels are adjusted to 1024*1536, the performance is not very good.

90年代亚洲女孩LoRA，基于SDXL1.0训练。

触发词：cn1990s

训练素材：36张90年代老照片，经过简单处理，去除了边框和文字，调整了像素。

基础模型：sd_xl_base_1.0_0.9vae.safetensors

泛化性能一般，建议设置强度为0.4~0.6。若把生成像素调整为1024*1536，表现不太好。