This is the LoRA model trained using Animagine V3.1 (SDXL).
I highly recommend using Animagine V3.1.
kanne_sdxl_v1.0.safetensors
Same model published on other platforms. May have additional downloads or version variants.