Type | |
Stats | 2,932 1,299 |
Reviews | (473) |
Published | May 17, 2023 |
Base Model | |
Training | Steps: 8,160 Epochs: 24 |
Trigger Words | janna \(league of legends\) |
Hash | AutoV2 223B35DD87 |
Lora trained on Janna from League of Legends
models used
anime: CoffeeMix, Anything and Everything, MeinaAlter
2,5D: MeinaUnreal - V3, GlassBones, Dark Sushi 2.5D
realistic: fantasticmix_real, henmix_real - v3.0
tags: janna \(league of legends\), 1girl
optional tags: elf, floating haar, blonde hair, elbow gloves, bare shoulders
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: orangemix.vae (for 2,5D pictures), vae-ft-mse-840000-ema-pruned (for realistic models), kl-f8-anime2, blessed2.vae (for anime models)
Sampler: DPM++ SDE Karras or DPM++ 2M Karras
Steps: 20 to 35
Clipskip: 1 or 2
Upscaler : img2img in comfyui using 4x-UltraMix_Smooth
CFG Scale : 5 to 9 (mostly 7 and 8)
LORA weight for txt2img: 0.5-1 (for the examples i mostly used 0.7)
Denoise strength for img2img: 0.4 to 0.7
If you want the exact same settings, just drag and drop the picture into ComfyUI
for the pictures i used comfyui