LoRA trained on Star Guardian Ahri from League of Legends
Update: added danbooru tags, completly different dataset lol, nai base model
This was my first LoRA, so it aged like milk. I used Youtube tutorials to create it and incorporated a combination of cosplay images and fan arts with blip caption lol. I really, really needed to update this xD
tags: star guardian ahri, 1girl, animal ears, hair ornament, detached sleeves, bare shoulders, skirt, magical girl, multiple tails
anime: GatchaMIX - v1.0, Zenith - v3, Yuzu - v1.1
2,5D: irRealitium, Aniverse - Aniverse V1.3, DarkSun, Dark Sushi 2.5D, ERA_09 - v1.0
realistic: fantasticmix - v3.0
better hands LoRA: EnvyBetterHands LoCon - beta2
negative embeddings: Fast Negative Embedding (+ FastNegativeV2) - 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, RealESRGAN_x4plus_anime_6B.pth
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
old Version was Trained with lyriel_v13 checkpoint
new Version was Trained with NAI checkpoint