Type | |
Stats | 849 1,030 |
Reviews | (116) |
Published | May 10, 2023 |
Base Model | |
Training | Epochs: 24 |
Trigger Words | kuniurban kuniurbanmask kunidefault kunimask |
Hash | AutoV2 DE17C1A9B1 |
Lora trained on Kunimitsu II from Tekken
Update:
new outfit and more possibilities to create custom outfits.
However, the original mask is less consistent, so I recommend using the old model for it
models used:
2,5D: ReV Animated, HELLmix, GlassBones, PsyGon Animerge - v1.1
realistic: fantasticmix_real, henmix_real
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, kl-f8-anime2, blessed2.vae (for anime models)
Sampler: DPM++ SDE Karras or DPM++ 2M Karras
Steps: 20
Clipskip: 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
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