v1
Trigger words: 3D, base, too many words may be ineffective. .
Size: 768×768, lower the resolution and open HD restoration.
lora weight: 0.5-1.5, 1 is recommended.
Lora is very random, or has a lot of imagination, and the content is 3D + base based on lora's imagination.
This is definitely not a good lora, but if you are interested, it may be a fun lora.
Function:
1. Describe the scene (it feels good)
2. To simply describe a person, the upper body + base (like a real statue) will be cut off and the description of actions, clothes, whole body and lower body will be invalid.
3. Used together with lora, it seems useless. It may become more realistic or more 3D.