I wanted to try creating a style that I like. After a lot of attempts, I trained and merged my own dataset with LoRa, and finally achieved this result. My computer deserves a break now. The Negative prompt has a significant impact on the image, too many prompt words can affect the outcome. The model is not perfect,to get a 2D image, you need to include '3D' in the Negative prompt.

AlchiliaLab

I wanted to try creating a style that I like. After a lot of attempts, I trained and merged my own dataset with LoRa, and finally achieved this result.The model is not perfect, to get a 2D image, you need to include '3D' in the Negative prompt.

alchilialab_V10.safetensors

AlchiliaLab_v1.0

I wanted to try creating a style that I like.After a lot of attempts, I trained and merged my own dataset with LoRa, and finally achieved this result.My computer deserves a break now.The Negative prompt has a significant impact on the image, too many prompt words can affect the outcome.The model is not perfect,to get a 2D image, you need to include '3D' in the Negative prompt.

Files with SHA256: 58a6327b1a02c5ecb212e62bdc10adcf6a9c4b53552aaf2ee02f07ad8e8f5e8e