Lora trained on Pyrrha Nikos from rwby
Master model 2,5D: ReV Animated (mostly used), OrangeChillMix (this one is also more realistic)
anime: DivineEleganceMix
i also used epi_noiseoffset with 0,7 weight
negative embeddings: embedding:bad-hands-5, embedding:ng_deepnegative_v1_75t, embedding:bad-artist, embedding:EasyNegative
postive promt for most of the colosseum pictures: ((best quality)), ((masterpiece)), ((realistic)), (detailed), (1 girl) women, pyrrha nikos a woman with long red hair wearing a brown corset and armor holding a spear and a shield, green eyes, full body, looking at viewer, solo, detailed background, (tribal background:1.2), colosseum, closed mouth, absurdres
postive promt for most of the Indoor pictures: ((best quality)), ((masterpiece)), ((realistic)), (detailed), (1 girl) women, pyrrha nikos a woman with long red hair wearing a brown corset, green eyes, looking at viewer, solo, upper body, detailed background, Ornate, Indoor, (Renaissance:1.2), (Vintage Windows:1.1), Royal Furniture, close up, closed mouth, absurdres
also try to keep this in mind if using ReV Animated:
Order matters - words near the front of your prompt are weighted more heavily than the things in the back of your prompt.
Prompt order - content type > description > style > composition
This model likes: ((best quality)), ((masterpiece)), (detailed) in beginning of prompt if you want anime-2.5D type
This model does great on PORTRAITS
Recommended settings:
Make sure you are aware on the usage instructions of LORA
VAE: orangemix.vae.pt (for 2,5D pictures), vae-ft-mse-840000-ema-pruned (for realistic models), pastel-waifu-diffusion.vae.pt and blessed2.vae.pt (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 0.8)
Denoise strength for img2img: 0.4 to 0.7
Trained with ReV Animated checkpoint
for the pictures i used comfyui