This Lora generates stereograms that can be viewed stereoscopically with cross-eyed viewing.
It genarates more fruitful outputs than my old SD1 ver. (I believe).
Trigger word: strgrm
Adetailer (face) is recommended.
Suitable weights is about 1.3～, depends on prompts, checkpoints, or other LoRAs.
Try to find the best value starting with a small value.
It appears that higher weights do not necessarily result in more stable output.
Ease of success depends on conditions. If it is a one-piece swimsuit, it is easier to get good output. It becomes more difficult when adding character LoRA. Try many times with trial and error with weights, etc.