Lora trained on Coven Ahri from League of Legends
Update: double the dataset, better base model (NAI), different kohya settings, trained on 768x768, lower steps better flexibility
Some pictures were done with ComfyUI since Civit finally supports its metadata
tags: coven ahri, 1girl, dress, detached sleeves, cleavage, multicolored hair, blonde hair, black hair, animal ears, pantyhose
optional tags: whisker markings, multiple tails, energy ball
You definitely need more tries on those ones since Stable Diffusion doesn't like counting
negative embeddings: Fast Negative Embedding v2
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(euler a for img2img upscaling) from time to time im also using DPM++ 2M alt Karras
Steps: 20 to 35
Clipskip: 2
Upscaler : GitHub - pkuliyi2015/multidiffusion-upscaler-for-automatic1111 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 and 1)
Denoise strength upscaling: 0.3 to 0.55
If you want to autocomplete the tags & trigger im using, I recommend using Booru style tag autocompletion for AUTOMATIC1111's Stable Diffusion web UI
For full body shots i recommend using: ComfyUI Impact Pack
or the a1111 alternative GitHub - Bing-su/adetailer