this lora is based on path to nowhere game, highrisemix and IDv2.10] Anime Illust Diffusion / 动漫插画扩散 and somethingAmazing checkpoint gives best results
Dark Age -v1 and XXmix_2.5D -v1.0 is also providing good results.
I also tried Yesmix-1.6 origin, it generates good faces but not good background details
MeinaMix-v11 and MeinaPastel-v6 is also good but with clearvae.safetensors and animevae.pt
for best results set lora weight to=1.0
cfg scale=10
sampling steps=30
in your prompt add "illustration","game background","official art" so that it looks similar to path to nowhere illustrations
Description
This version is trained on civitai using sdxl with better details
all the images are generated using JuggerMagine XL (https://tensor.art/models/690154360874918446) and animagine-XL version 3.0
CFG scale-5 to 7
Sampling steps- 20 to max
lora strength-0.8-1.0
FAQ
Details
Files
Available On (3 platforms)
Same model published on other platforms. May have additional downloads or version variants.