Last updated - 22/06/2026
Note - As the supporting articles for this LoRA cover all of MGE's monsters, they contain some NSFW content, so you'll need to log into Civitai.Red to see them.
(The first Anima version has been released separately due to the difference in architecture, for those interested.)
I now have a Ko-fi for those that'd like to support my projects. While nothing there will remain exclusive, you'll get more substantial progress updates each month, and access to my own non-AI images at higher tiers.
Model description:
This is an attempt at an all-in-one lora for generating the monster girls of MGE, which allows for extensive customisation and feature-mixing. Refer to these articles for more information. Not everything works perfectly, so good results may require a bit of experimentation at times.
For updates on future versions, I generally post what I'm working on in my profile's announcement tab. It'll probably be MGE-related, but in addition to this one I occasionally make standalone concept-loras for monsters that would benefit from having one.
Background info:
"MGE" Refers to the monster girl encyclopedia book series by Kenkou Cross. For more information, see the MGE wiki here.
Additionally, while I'm not affiliated with them, the wiki recently opened donations so you can donate here if you'd like. They have a $12/month upkeep cost apparently, so I'm sure they'd appreciate the generosity. Be sure to mention where you got the link from too, for the model's sake :).
"Non-entertainment" usage:
Any "commercial projects" must be transformative, so no posting your results to Patreon and asking for money. Using it as part of a larger project is fine though, and I'm happy to help if you reach out with questions, or simply to tell me an interesting use-case you have. Bear in mind that if you're creating stuff linked to MGE specifically, they must also adhere to the creator of MGE's terms here.
Actual artists:
If any artists dislike the idea of their stuff being used for this (and you've drawn things from MGE), let me know and I'll avoid your work. I haven't done this preemptively due to the huge number of dead source links in the archive, but if you reach out (with some proof of identity to avoid trolls) I'll remove any of your content from my dataset/s.
Reuploads:
If you reupload this, please link both this model page and my taglist for ease of use. I'd also appreciate you letting me know where you've uploaded it, so I can keep track of where it's being used.
Description
FAQ
Comments (40)
Is there a downloadable taglist or list of trigger words for this ~0.5GB lora?
I've had a request for a Qwen version, so react to this comment if you're interested, and I'll look into training one if there's enough demand.
Also, feel free to suggest other base models below this, and react to the ones you'd like a version for. I've been stuck to Illustrious for a while now, and haven't tried many newer models, so it'd be interesting to see what else people are using nowadays.
Small update to this, Civitai doesn't currently support Qwen for training. I'll train a version if/when they add the option, but unfortunately I can't do so right now.
Do you have any request for ZIT? would love to see this on ZIT.
@Wheeman I'll likely do a ZIT version with V7.1, It being caption-based instead of tag-based means it'll require a full rewrite of my dataset, so more work than others. I've had a couple of requests, but the majority of people seem to still be using Illustrious, so I didn't think it'd be worth going back to V7.0 having already started testing a few updates post-release.
@Blizzard264 I understand, yesterday I managed to do Snake girl with ZIT and omg what a pain... I had to include mermaid with long tail and more weird things to nake it right.
anyway, take your time. happy to wait.
Just a little Question, ive been in love with your lora since day one i discover it back in the days of pony. Lately in the last version of illustrious ive noticed that using the lora with Wai-illustrious 15.0 and 14.0 and Janku 5.0 (this are the models i use more often) tends to leave a box in the face, like a box with a slight different color, sometimes is very noticeable sometimes not so much.
To be specific it happens when i use Adetailer for face, i managed to reduce the issue setting the Adetailer 'Inpaint denoising strength' to 0.2. it helps but it doesnt fix the problem completely. So i was curious, why does this happens, if the issues has been noticed before, and if theres a solution, or maybe its Adetailer who has the problem and other options are being used.
thanks for your time n.n
@Razuryam It's not an issue I've noticed, but I'll look into it. I use WAI-Illustrious 14 too, so I'd assume it's an adetailer-related issue, especially if reducing the denoising strength helps the issue somewhat. I haven't had issues with adetailer myself though either, and I've used it a fair amount, so I might need some examples.
@Blizzard264 mhm! i thought the same and thats fair, how should i show u examples? do i make a post with the box i mentioned. it'll maybe take me a while, but once im free from work, ill do a compilation of some examples and show u.
@Razuryam Yeah posting's probably the easiest way, although if you'd rather keep them private I don't mind using Discord or something too. Try to get a couple of examples both with and without adetailer enabled so I can see exactly what changes, and I'll see if I can reproduce the issue.
@Blizzard264 Actually discord could be a good option, at least it makes communication faster and easier imo, If youre okay with it of course. Were can i find your discord to send you a message?.
@Razuryam It should be linked under my profile somewhere. I'll send you a link in DMs though, because a couple of people said they couldn't find it either.
@Blizzard264 yeah sounds great, found the little icon, but it just redirects to discord main page.
Check your VAE - I had same problem with PureColor. Each inpainting results in slightly darker and pale but noticeable colors
I have same problem, any luck on this one?
@Jackousla Yeah, try turning up the "inpainting mask blur" setting, should be under the "inpainting" tab of adetailer.
@Jackousla Yeah just like blizzard suggested, we found the solution by turning up the "inpainting mask blur"
This is my favorite lora ever
best lora
Oh, I missed the release of version 7.
Well, now there's a reason to redraw my favorite monster girls. Ver. 6.1 was really good.
7.0 is not up to 6.1
W model pull
React to this comment if you're interested in an Anima version.
I'll have to wait for Civitai to add it as a training option regardless, just wanted to get some early feedback for if/when they do.
Honestly i think the transition to anima is just the natural evolution for this amazing Lora. Ngl im a little scared because i dont know if i can run anima tho xD. But ill hope i can.
@Razuryam I still don't know much about it tbh, haven't tried it yet, I've just been hearing it mentioned a lot lately. I am a bit sceptical as the hype reminds me of Z-image, which didn't really go anywhere afaik, but I guess we'll see what happens.
@Blizzard264 Honestly same, i know little to nothing about Anima, but mostly got my attention cause i saw that WAI-NSFW creator got a version for Anima, ive noticed that the models seems to weight less than Illustrious ones.
Beside that ive just seen very good art from some people using Anima in the Image Gallery... and it kinda made me want to try... but im a little hesitant because i need to install a different stable diffusion than the one im using... but maybe is for the best since mine is not getting updates anymore :')
@Razuryam anima is slower yet lighter than sdxl if im not mistaken. almost all of my work is from anima tho im new here
@Ashendm Is it tag-based, or another description-based model? I feel like that's the main thing a lot of the "SDXL killers" have been missing, as I'm under the impression the latter is more awkward to work with in terms of dataset and/or prompt writing.
@Blizzard264im not saying this to get attention to my images trust me its a good mix , you can go either way BUT checkpoint and training way plays a huge role too. check out my images for better understanding or here are some obvious images to see the prompts
https://civitai.red/images/131374812
https://civitai.red/images/131362873
https://civitai.red/images/131316237
https://civitai.red/images/130871326
https://civitai.red/images/131020315
https://civitai.red/images/130564924
im pretty new her so images are not super great BUT illust and sdxl would never give results closer to this with this trash prompt
@Ashendm No worries lol, but yeah it seems promising, I like the idea of a more open-ended system. I guess I'll just have to test whether my current tagging system works when it's added to the site.
basically a checkpoint for the size of 2 full size loras. great work dude
Small update again: I've started trying to train an Anima version but I haven't been able to get good results with it yet. I believe the main issue is with AI Toolkit, as my prior tests with Illustrious had the same issues, but it's unfortunately the only trainer available for Anima on-site. I'll give it a few more goes, and I'll probably release whatever I get as a test version to demonstrate what I mean, but a proper Anima version may have to wait.
My next model will be V7.1 of this regardless, so don't worry if you're an Illustrious user, I'm still working on that too. I just wanted to update those waiting on an Anima version, with how much interest people have shown in it thus far.
Do u think to tried make anima version?
I think it better when monster girls have too many characteristics, Anima does a better job of creating more detail for the girls.
@BeeZ333 Currently working on it, just taking a while due to Civitai's ongoing issues, I've had 7 training runs fail in a row so far... I'm not sure I'd be able to match the performance of the Illustrious versions right now though, regardless, as they're only offering AI-Toolkit-training for Anima, which unfortunately doesn't work anywhere near as well as Kohya.
@Blizzard264 Oh, that sounds inconvenient...
I hope you have good luck with it <3
Keep trying!
@BeeZ333 Alright, the test version's out if you wanna give it a try, might be a bit hit-or-miss but some things seem to perform better.
Another update: The first Anima version is now released. Let me know what you think, I've been getting mixed-but-promising results thus far, but I haven't had the chance to do a full testing run yet.