
My first try of a SDXL Lora of the character Ayanami from the mobile game Azur Lane.
Lora weight of 1.0 is recommended.
Trigger word is "ayanami_azur_lane", mostly it doesn't require more than that.
Peripheral prompts: blonde hair, long hair, ponytail, hairpin, headgear (just refer to the sample images)
I highly recommend using Animagine XL 3.0 as I found the lora performs here the best (obviously this is subjective).
Sample images are made with Hires. fix [R-ESRGAN 4x+ Anime6B; 18 steps; denoise: 0.2~0.3; upscale x2] and ADetailer face_yolov8n model [prompt: "<lora:Ayanami-ayanami_azur_lane:1>,ayanami_azur_lane,amber eyes"; denoise: 0.2~0.3]
Feel free to leave a comment no matter whether positive or negative as maybe I can learn something and improve the lora in future.
描述:
initial release
训练词语: ayanami_azur_lane
名称: Ayanami-ayanami_azur_lane_final.safetensors
大小 (KB): 223111
类型: Model
Pickle 扫描结果: Success
Pickle 扫描信息: No Pickle imports
病毒扫描结果: Success