For the best result for my LoRA, u must using checkpoint Animagine-XL V3.1
and my LoRA, or u add other LoRA for more experients, but less is simple. :)
My Best Settings for this LoRA:
weight: 0.8-1
sampling steps: 25-33
VAE: vae-ft-mse-840000-ema-pruned
sampling method: DPM++ 2M Karras or DPM++ 3M SDE Karras
CFG Scale: 7-9
clips skip: 1-2
Upscale: R-ESRGAN 4x+ Anime6B
Step : 20-25
Denoising strength: 0.3-0.5
If u like this LoRA, SUPPORT ME