https://civitai.com/models/103697?modelVersionId=111080
Uploader note:
in terms of Commercial use, only selling images is allowed, both paid generation as well as selling the model or merges isnt.
Author Description:
LoRA trained on Fino Bloodstone from Yuusha ni Narenakatta Ore wa Shibushibu Shuushoku wo Ketsui Shimashita (I Couldn't Become a Hero, so I Reluctantly Decided to Get a Job I Couldn't Become a Hero, so I Reluctantly Decided to Get a Job)
requested by: @RiderBond1949
Version 2 I experimented with a learning rate of 0.0001 and used six outfits that are closer to the original but less flexible and more overtrained. Version 1 is more stable, so I would generally recommend using it.
models used
anime: Cetus-Mix - V4, YabaL_Mix - V5, MeinaPastel - V6 ( Pastel ), Sweet-mix - v2.1
2,5D: Animesh - Animesh-Full V1.5, PastelDiffusedMix - V2.1
negative embeddings: bad_prompt, negative_hand, ng_deepnegative_v1_75t, bad-artist, 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(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
