Character model based on all currently existing 52 images of Macha, an Eldar/Aeldari Farseer from Warhammer 40k videogame series Dawn of War. Most of it is the lovable goofball version we all know and love courtesy of lewdanon, but perfectly capable of doing a more seerious version too.
Model is trained on NoobAI ε-pred 1.0 and is untested on base ILXL.
Usage
Basic
Works well anywhere between a strength of 0.8 and 1.0
Base trigger is farseermacha
Note: If dropping below 1.0 you may need to specify green eyes.
Outfits
Her armor is more or less reproducible with:white armor, green robe or green armor
Extra
In the interest of openness and community-helping, I have uploaded the dataset along with the model, this includes: images, captions, and the training json (for kohya_ss).
Description
trained on 52 images on NoobAI ε-pred 1.0
FAQ
Comments (6)
Man, you are amazing! Please, adapt more characters for NoobAI, this is really new META in art models ;)
After some fiddling, it definitely seems better than both Pony and Illustrious at almost everything, just needs more momentum from other users.
Would be nice if Civit could finally give it its own model category too 🥲
@DrA_263 However it's sad Civit somehow doesn't support v-pred models, and NoobAI author just updated it yesterday with a new version. Instead, can use NoobAICyberFix.
@shaerahall715 Mmm... I don't know if the v-pred models are very good though. I made a post about it on the NoobAI model page, but detailed backgrounds seem to be basically a no-go - they always come out as blurry messes.
Is it possible to implement as update an accurate canon eldar weapons for her?
Would have to be a whole separate model, but sadly I don't think there's really much training data for witchblades or singing spears, etc.
Maybe if Games Workshop could stop making space marine content for 5 minutes and let us see the other factions, we might get more fanart of them 😭
Details
Files
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.