mggirl SDXL -test version 2

This checkpoint training dataset primarily focuses on portraits and is suitable for images where the face occupies more than 1/3 of the frame.

It uses the simple trigger word 'mggirl', and the style of the images leans towards a nostalgic vintage photograph style.

Please try to run the model with a standard size of 1024x1024.

This is still a version under testing, so after running it, please make sure to send the results back to me. I will take the time to continually optimize it. Thank you!