Type | |
Stats | 766 |
Reviews | (116) |
Published | Jun 5, 2023 |
Base Model | |
Training | Steps: 1,950 Epochs: 15 |
Trigger Words | coven lissandra |
Hash | AutoV2 6302E0F977 |
LoRA trained on Coven Lissandra from League of Legends
requested by @Cinders
models used
anime: YabaL_Mix V3(image 5), MeinaAlter, CoffeeMix(image 6)
2,5D: DarkSun(image 1,2,3,4,7), HELLmix, GhostMix, MeinaUnreal - V3
realistic: Paragon V1.0(image 8,9,10)
tags: coven lissandra, 1girl,
optional tags: helmet, white dress, detached sleeves, blonde hair, helmet over eyes, 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 4x-UltraSharp(for 2,5D & realistic), 4xAnimesharp(for anime)
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 full body shots i recommend using: ComfyUI Impact Pack
or the a1111 alternative GitHub - dustysys/ddetailer