Lora trained on Emerald Sustrai from RWBY
models used
anime: MeinaUnreal, ChinoMix
2,5D: Lyriel, fantasticmix_real, ReV Animated
classic outfit tags: emerald sustrai, 1girl, dark-skinned female, green hair, bob cut, classic outfit, red eyes, halterneck
optional: bare shoulders, holding weapon, gun, holding gun, white pants
atlas outfit tags: emerald sustrai, 1girl, dark-skinned female, green hair, bob cut, atlas outfit, cropped jacket, white pants, red eyes
optional: zipper, white cropped jacket
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: 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 (0.7-0.8 recommended)
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