Type | |
Stats | 501 |
Reviews | (69) |
Published | Jun 2, 2023 |
Base Model | |
Training | Steps: 4,515 Epochs: 21 |
Trigger Words | dottore |
Hash | AutoV2 9CB31740DC |
LoRA trained on Dottore from Genshin Impact
requested by: @sayunara
models used
anime: YabaL_Mix V3(image 6,7,9), MeinaAlter(image 1), PrimeMix - anime(image 12)
2,5D: MeinaUnreal(image 2,4,5), HELLmix(image 8), DarkSun(image 3,10,11)
tags: dottore, 1boy,
optional tags: light blue hair, fur-trimmed coat, white coat, male focus, blue shirt, o-ring, chest harness, collared shirt
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