illustriousXL/NoobAI Stabilizer
TD;LR:
This LoRA can make the character pose and style more stable and natural.
Recommended strength: 0.4~0.8. Theoretically, can go up to 2 if you want. (rank/alpha=2).
No trigger words needed. However, highly recommend to delete all negetive prompts for better detail.
Compatible with NoobAI v-pred model.
Can "fix" messy noisy image when using v-pred model on Civitai online generator. See below.