LoRA of singer Lizzo, trained on 21 images. The trigger word is lizz0.

The 128dim works best at 1.2-1.3 strength and the ??dim is best at 1.3 strength. While I didn't use her weight in any of my training tags, during image generation I found that I needed to add something like "she is a large woman" or it tends to make her thinner.

Like my work? Please consider leaving a tip to help fund future LoRAs or commission a custom LoRA for yourself. Higher quality versions of my LoRAs are also available for a monthly donation: https://ko-fi.com/artfulgenie.