LoRA trained on Porcelain Kindred from League of Legends

requested by: @BangBangFukanawa

models used

anime: YabaL_Mix V3(image 3,5,7), MeinaAlter(image 1), PrimeMix - anime(image 10), DrizzleMix(image 1)

2,5D: CamelliaMIx_2.5D - V2(image 8,9), HELLmix, DarkSun(image 2,4)

tags: porcelain kindred, 1girl,

optional tags: helmet, blue hair, dress, blue pants, knee boots, covered eyes

negative embeddings: embedding:bad-hands-5, embedding:ng_deepnegative_v1_75t, embedding:bad-artist, embedding:EasyNegative

Recommended settings:

• Make sure you are aware on the usage instructions of LORA

• VAE: vae-ft-mse-840000-ema-pruned (for realistic models & 2,5D), kl-f8-anime2, blessed2.vae (for anime models)

• Sampler: DPM++ SDE Karras or DPM++ 2M Karras

• Steps: 20 to 35

• Clipskip: 2

• Upscaler : img2img in comfyui using 4xAnimesharp

• CFG Scale : 5 to 9 (mostly 7)

• LORA weight for txt2img: 0.5-1 (for the examples i mostly used 0.8)

• Denoise strength for img2img: 0.4 to 0.7

for the pictures i used comfyui