/mg/ - AI Monster Girls Thread

/mg/ - Monster Girls

Main monster girl board

Mode: Reply

Name
Subject
Message	Max message length: 20000
Files	Drag files to upload or click here to select them Max file size: 30.00 MB Max files: 5
More
E-mail
Password	(used to delete files and postings)
Misc

Remember to follow the rules

[ Return / Catalog / Bottom ]

AI Monster Girls Thread Anonymous 07/11/2024 (Thu) 11:26:06 No. 11282

I've been taking a look at AI again recently and the technology is very impressive. Seems most decent images these days come from an SDXL model called Pony Diffusion. https://civitai.com/models/257749/pony-diffusion-v6-xl Yes, it has furry shit but you can turn it all off since it comes with built in negative prompts for entire parts of the dataset. Despite being named after ponies 50% of the dataset was anime, and it's apparently the best or one of the best local models for generating it right now. It intuitively knows how to do sex, blowjobs, footjobs, etc without any loras. >What is a lora? Just think of it as an addon that adds additional features to a model it is trained for. (You can't use loras that aren't compatible with your model). I use Automatic1111 which is buggy as shit but has a lot of good extensions and I've put far too much time into it. https://github.com/AUTOMATIC1111/stable-diffusion-webui/wiki/Install-and-Run-on-NVidia-GPUs Honestly, there are a few extensions that basically force me to stay now, and let me tell you about them: https://github.com/zixaphir/Stable-Diffusion-Webui-Civitai-Helper This labels everything you download from CivitAI and allows you to load in the "best prompt" provided with the lora, along with inserting all the model's trigger words and giving display images to each of your loras. https://github.com/Mikubill/sd-webui-controlnet Lets you use an image as direct reference for generating, so you could maybe even get tricky stuff like tail pussys or slime girls to work if you do it right. I'm not quite familiar with it yet but it's already performed really well. I had the best experience with anime canny and lineart models. You need to download separate models for it and put them in your a1111 folder's models/ControlNet folder. For SDXL, you can find those Here: https://civitai.com/models/136070?modelVersionId=267516 And here: https://huggingface.co/kohya-ss/controlnet-lllite/tree/main Finally, by far the most important one, ADetailer. https://github.com/Bing-su/adetailer It automatically detects faces in your picture and inpaints them to keep them from looking terrible. You can provide custom prompts for finetuning. Things to note: I recommend at least having a minimum of 6GB vram and an RTX 20 series or greater gpu. If you have below 9GB of vram, you should probably use xformers. Look at the automatic1111 wiki for how to do that. You should also check out https://github.com/AUTOMATIC1111/stable-diffusion-webui/wiki/Optimizations if it runs too slowly.

Anonymous 06/22/2025 (Sun) 16:39:42 No. 27031

>>27005 >I use Reforge, which is basically identical UI-wise to A1111 as far as I can tell but through some witchcraft is like 30x faster. That's precisely what would fix my problems. Though Comfyui's node system also seems really powerful. What OS and specs are you running those on?

Anonymous 06/22/2025 (Sun) 16:41:25 No. 27032

>>27005 >Reforge https://github.com/Panchovix/stable-diffusion-webui-reForge Is it this thing?

Anonymous 06/22/2025 (Sun) 17:04:16 No. 27040

>>27005 >your third pic >Commissar red-chan >Stop that! I am not nice! I'm threatening! >I'll send you to the Goo-lag!

Anonymous 06/22/2025 (Sun) 17:16:59 No. 27043

>>27040 The model almost got the jap right, it wrote "natte, nadde" (though the "na" syllable is quite broken), and petting is "nade nade" in jap.

Anonymous 06/22/2025 (Sun) 22:40:29 No. 27112

Anonymous 06/23/2025 (Mon) 04:19:17 No. 27134

>>27031 >What OS and specs are you running those on? Win10, RTX 4060 (8GB vram lol, didn't do any research when I bought this thing, just picked a random budget-ish card in an emergency). Takes 20 secs to generate a draft image, about a minute with hires fix and adetailer, usually. With A1111, it could take upwards of 15 minutes for a single image depending on how spiteful the model was feeling, which was far from ideal. Never used comfy so I can't comment on it. >>27032 Probably, looks about right. >>27040 I generated her for a SillyTavern card I'm working on, where a militant neonazi slime girl declares (You)r home property of Deutschland. She keeps a framed photo of Hitler in her body. I am stealing Goo-lag.

Anonymous 06/23/2025 (Mon) 05:58:36 No. 27144

>>27134 >2nd pic Finally, the cuckquean memes have been put to rest

Anonymous 06/23/2025 (Mon) 07:24:04 No. 27159

>>27144 Tamamo is the one sucking the tip and getting most of the semen, cucklice can only lick the shaft and lick off excess cum

Anonymous 06/23/2025 (Mon) 21:48:09 No. 27236

Post your specs, models and gen times everynyan :3 No, really, I want to get a rough idea of benchmarks of different gpus and alike, I'm trying to gather as much info as I can before buying a pc for this.

Anonymous 06/26/2025 (Thu) 05:40:01 No. 27573

>>27032 Been thinking about making a new thread op or at least a guide about how to install AI locally. A lot has changed recently so I think it's about time. Tell me if you're interested so I have enough reason to bother lol. I actually don't recommend ReForge anymore because it's ceased development. I'd just use regular Forge- though, the main creator of that stopped working on the repo a long time ago, though they didn't officially retire. Forge still gets semi-regular updates from contributors. I would not use A1111 original. I would hardly call it functional anymore. Random bullshit errors, slow as hell, not having support for new things like FLUX. Speaking of FLUX, it is finally becoming usable for NSFW. Someone is making an NSFW version of it, "Chroma". It's in early development, but already has support in Forge. You'd probably need a 4090 or 5090 for it though. At least a 3080 or 3080TI.

Anonymous 06/26/2025 (Thu) 05:49:53 No. 27574

>>27236 Main thing that matters for reasonable performance is essentially -Must be NVIDIA -Must have 16GB VRAM. If you do not have VRAM, things will slow down for you *exponentially*. With VRAM, unless your GPU is absolutely lowend, it will finish eventually. If you absolutely must, get a 12GB vram card, but you won't have much room to run improved/bigger models in the future. With low budget, I'd probably get a used 4070 TI SUPER or 5060 TI (IMPORTANT: 16GB edition, NVIDIA has a scam 8gb edition) If you have that, you can probably at least get something working well. I can generate 8 hires fix/upscaled images with Illustrious/SDXL models on the 5090 in about a minute and a half, ADetailer processing included.

Anonymous 06/27/2025 (Fri) 18:01:10 No. 27761

>>27236 i just use novelAI. They released 4.5 compatible with vibe transfer, and will experiment a lot with that over the weekend for posing and such.

Anonymous 06/27/2025 (Fri) 18:03:40 No. 27762

>>27761 Lol, did not notice major error, fixed. My pc is also nothing special in terms of performance, which is why i didn't include it. I doubt it could run local efficiently.

Anonymous 06/27/2025 (Fri) 18:26:00 No. 27765

At the risk of spamming (who cares ded thred amirite) the new capabilities are fuckin wild. Two lewder images ive been able to get today.

Anonymous 06/27/2025 (Fri) 21:29:17 No. 27784

>>27765 Made another lora, this time frfr's style. Tried to get his old style from the original MGQ vns for nostalgia, didn't quite work out. I think it's meshing the style of some of his newer work with all of that. Might try training one on just the old mgq cgs from frfr. Still turned out very will with my tests on an Ilias lora.

Anonymous 06/27/2025 (Fri) 22:04:53 No. 27791

>>27784 based and angelpilled

Anonymous 06/27/2025 (Fri) 22:13:27 No. 27792

>>27784 That's really good, only thing that stands out is the weird messed up nigger torso on second pic

Anonymous 06/27/2025 (Fri) 22:39:57 No. 27796

>>27784 did you post a full step by step for dumb asses yet?

Anonymous 06/27/2025 (Fri) 23:41:10 No. 27799

>>27796 For making loras? There's a bit of a tangent I go on about it up there, it's not exactly step-by-step but it has a lot of information. I didn't bother to make a guide because I've been a bit busy and no one really replied or seemed interested.

Anonymous 06/28/2025 (Sat) 00:27:13 No. 27803

>>27799 No, for all of it. I stopped paying attention a few years ago and I think the new meta is comfyUI or something.

Anonymous 06/28/2025 (Sat) 01:15:45 No. 27807

>>27803 The OP is still somewhat relevant but instead I'd use Forge or maybe try SwarmUI, but I haven't used the latter personally. I don't recommend ComfyUI for anyone trying to do it casually since it's literally not even a UI. You can think of it like Scratch (the coding block programming language) but for generating images. Or, flowcharts. It is definitely very well maintained, though. I'm running into annoying issues with Forge. It's barely compatible with the 50 series. Or maybe it's just collapsing in general. Or perhaps it's NVIDIA's drivers. Who knows.

Anonymous 06/28/2025 (Sat) 12:34:29 No. 27853

>>27765 >At the risk of spamming >spamming >4 pics in 3 posts >all relevant If anything, I think you should post more. >the new capabilities are fuckin wild. How's it different from before? Show some screencaps of the process too, if you don't mind. >>27784 >Made another lora, this time frfr's style. >Tried to get his old style from the original MGQ vns for nostalgia, didn't quite work out. Looks plenty nostalgic to me. >I think it's meshing the style of some of his newer work with all of that. I think the lines are too clean, his outlines were always a bit sketchy. >>27792 Looks like a dark shirt, notice the edge near the crotch.

Anonymous 06/28/2025 (Sat) 12:39:54 No. 27857

>>27573 I think a new guide would be good. >>27807 >I don't recommend ComfyUI for anyone trying to do it casually since it's literally not even a UI. You can think of it like Scratch (the coding block programming language) but for generating images. I tried comfy and it only looks scary at first. You can literally just load up the default workflow and start typing prompts like normal. Also its easy to install, at least on Win11, the installer basically does everything for you. And its fast. >>27807 >I'm running into annoying issues with Forge. It's barely compatible with the 50 series. Or maybe it's just collapsing in general. Or perhaps it's NVIDIA's drivers. Who knows. I'm thinking of buying a 5060 Ti 16gb, so what happens with it? You're the 5090 anon?

Anonymous 06/28/2025 (Sat) 13:47:49 No. 27873

>>27807 There is just stuff I don't really understand like checkpoints and vaes. Swarm is remarkably obtuse about this.

Anonymous 06/28/2025 (Sat) 18:15:48 No. 27912

>>27857 I think there have been like, at most, 6 people on this thread ever. If you see an in depth explanation about local Stable Diffusion anywhere in this thread, it's probably me. I'm thread op, 5090 anon, basically any major tech post. Stopped posting for half a year then came back when I saw activity on the thread. I guess I'll get around to making a guide later, but I'm exhausted right now from doing other things so probably not for a while. I guess I'll try Comfy, I'm just extremely reluctant to lose all my extensions like ADetailer and re-import my models. I also heard Comfy was more memory efficient. I think I WILL take a look at Swarm later, simply because the main backend of it is not the piece of shit that is Python. Hopefully, it would be more stable. I'll tell you if I figure it out. >I'm thinking of buying a 5060 Ti 16gb, so what happens with it? Ideally, it would eventually get official support for the new cards. It should still probably be fine to use it with Forge. Some people on the Forge github in issues and discussions have made their own patches for some functionality with the 50 series. There are bugs, but it can at least generate images. And if not, you can work out Comfy or SwarmUI.

Anonymous 06/28/2025 (Sat) 19:41:33 No. 27930

>>27912 >I guess I'll try Comfy, I'm just extremely reluctant to lose all my extensions like ADetailer and re-import my models. I dunno your workflow, but just trying it out shouldn't break anything else, its not like you're moving the model/lora files, right? >>27912 >Some people on the Forge github in issues and discussions have made their own patches for some functionality with the 50 series. There are bugs, but it can at least generate images. And if not, you can work out Comfy or SwarmUI. Yeah, I wanted to know what these bugs/issues were. You said you were running into them, so I hoped you would elaborate.

Anonymous 06/28/2025 (Sat) 19:42:33 No. 27931

>>27912 I used swarm. But the main thing is there is so much division going on right now. But I would say comfyUI is the most commonly mentioned thing even if it's not the easiest to use. whether that's a bias because it's hard to use is hard to know.

Anonymous 06/28/2025 (Sat) 21:46:55 No. 27948

>>27930 Running into an issue some other people are having where switching the checkpoint/model after booting can cause forge to generate only black squares or distort the model output (same seed generates a different image than before, usually with a different style or worse quality) It also might just trigger randomly, but usually I can generate 16-32 images so it's not intolerable. It's annoying, requires a full relaunch. I also just question whether or not the performance is even as good as it should be for my card, though I could just be imagining it. Seems almost slower in some aspects, but it could be that I've just adjusted to the new speed. I tested SwarmUI, it apparently uses Comfy as a backend, fortunately, it does let you configure the model folder so that wasn't an issue. Install was seamless. I think it's a bit slower than Forge for me, though. Something like 1.5x to 2x faster. Not having the same issue I had with Forge though since Comfy already has official support for 50 series.

Anonymous 06/28/2025 (Sat) 22:55:17 No. 27952

>>27853 It's not different as in a new function, it still works the same. A mix of vibe transfer, inpainting, and prose prompting. The 4.5 is just a little better at posing, object and background generation, and direction. Im still figuring it out though. the big update was 4.5 finally fucking getting vibe transfer, which is the NAI Mini-Lora imitation where you can upload reference images and scale their info extracted/strength individually.

Anonymous 06/29/2025 (Sun) 14:15:18 No. 28030

>>27574 >-Must have 16GB VRAM. tfw cucked by 12GB 5070

Anonymous 07/03/2025 (Thu) 02:07:32 No. 28643

Anonymous 07/03/2025 (Thu) 12:23:13 No. 28676

Anonymous 07/04/2025 (Fri) 05:43:20 No. 28782

>>28676 >that sausage picrel >cakesucc Very nice. The cream splashes look a bit off, but the idea is nice. Overall these 2 pics is what I want to see more of: actually using the tool, not just generating another simple POV fuck scene with 0 composition. Keep it up.

Anonymous 07/04/2025 (Fri) 14:05:26 No. 28839

And I'm still using a model that's like a year out of date.

Anonymous 07/04/2025 (Fri) 15:25:30 No. 28856

Anonymous 07/04/2025 (Fri) 16:49:44 No. 28886

Started to mess around with NAI again and honestly I'm kinda here for it

Anonymous 07/04/2025 (Fri) 17:57:29 No. 28909

>>28886 Like a moth to the flame, more will come. >>28782 Out of every one of those, cakesucc was by far the hardest to do. I had to manually edit on the gloops and the inpaint would continuously not recognize them, so i'de low-strength roll them, edit slightly, do again, etc. I also tried to do crumbs of the cake but same problem. A lot of AI is like directing a highly skilled blind person. Another fun fact, the frog girl wasn't that bad, but the fucking sausage/hotdog was so annoying because AI sees hotdog, and is like, oh yeah lets drizzle mustard and ketchup all over that bitch every time no matter what. I ended up fusing corndog (because those are usually on sticks) and sausage to finally get a useable standin, lmao.

Anonymous 07/04/2025 (Fri) 17:58:29 No. 28910

>>28909 first time i did (you)'s wrong, second time i forgot image, galaxy brain levels of retard. here you go, lol.

Anonymous 07/04/2025 (Fri) 20:10:48 No. 28939

>>28886 wow, first pic is really something else

Anonymous 07/04/2025 (Fri) 22:11:51 No. 28943

>>28676 >>28909 You used NAI for that? >>28839 What model? Also year old sounds like brand new to me. >>28939 Lol, lmao even. Unless you're sarcastic.

Anonymous 07/04/2025 (Fri) 22:20:42 No. 28945

>>28943 >Unless you're sarcastic. It has the AIslop shine but this actually isn't a problem with Demon's signature blue skin. other than that it's very good, only other weirdness I can spot as a layman is her right hand that's gripping the nipple and it's too small. The womb tattoo, dick, pussy and nipples with their piercings are all perfectly normal and that's impressive

Anonymous 07/04/2025 (Fri) 22:38:41 No. 28948

>>28943 >Lol, lmao even. Unless you're sarcastic. Man, who pissed in your soup?

Anonymous 07/04/2025 (Fri) 22:39:50 No. 28949

>>28943 Yeah, i use NAI and MS paint for my gens.

Anonymous 07/04/2025 (Fri) 22:46:53 No. 28950

Also, have this one which is a retarded joke from a different thread/board with no context whatsoever.

Anonymous 07/04/2025 (Fri) 22:47:27 No. 28951

>apologies for the non-monster pics Turns out I do have some free tries with NovelAI. 30 to be exact. First I experimented with some gyaru loli that was recommended as an "inspiration". I found it hilarious how the free trial was censored, but also tried to obey the lewd prompts as well, resulting in... well, you see what. Then I tried generating my rabbit character... Then I looked at my art... And it left a pit in my stomach. A simple mediocre drawing takes me like 90 minutes, the computer can do so much better in mere seconds. Sure, it needs polishing, but after polishing you have a reliable, insanely fast way to create any drawing. Whereas by hand every single simple drawing takes literal hours. Sure, you can talk about "soul" or whatever, and I do think my drawings have it, but so do the gens, so what? What can justify over 1000x increase in speed, all the while ALSO increasing the quality? I never even made a single cent with drawing, yet I still feel replaceable now. I thought I got over AI, and accepted "no rules just tools", but seeing it firsthand be better than me at drawing my character still had an effect. Damn...

Anonymous 07/04/2025 (Fri) 22:55:19 No. 28953

>>28945 >>28948 >messy hands >generic ai style >1girl, pov >no composition Cmon now. The tools have evolved, we're past that. >>28949 >Yeah, i use NAI and MS paint for my gens. >MS paint That's what I'm talking about, actually using the tool! Though I feel obligated to shill Krita, its very good, even if you're not into drawing. Recently it also implemented some live ai drawing tools, they look like absolute black magic, tho you need a graphics card for that. I saw some dude run it with 6gb of VRAM, though I can't make any promises. https://www.youtube.com/watch?v=nkVjXjzs4Z0 https://www.youtube.com/watch?v=Ly6USRwTHe0

Anonymous 07/04/2025 (Fri) 23:23:53 No. 28955

>>28953 Yeah, that looks to be kind of what i do but all rolled into one program. I wonder if it can link up to a style type via reference images or something. I usually like to generate several images of a character in different poses and slap them on as reference images. 1. imagine scenario 1a. if i already have characters made, pick reference images 2. Generate images of scenario and modify prose until i get a mostly acceptable pose or lighting or whatever i am looking for 3. adjust prose and generate variations of the "close" image 4. Draw and inpaint to blend the drawing without losing coherency The below images are a human maid and a dark elf maid (i have genned an absolutely absurd number of these two in solo and duo pictures). These aren't perfect but they are an example of keeping a mostly coherent style and characters/proportions. If Krita can do that and i can keep the styles i want from the base images i gen on NAI, that'd be fuckin' cool. I'll look into that for sure.

Anonymous 07/04/2025 (Fri) 23:54:05 No. 28961

>>28955 You can do consistent styles and characters using loras, though if you're making absurd amounts of OCs then training them all would be a pain. I dunno about a "reference" function in local programs, though I saw one "pseudo lora" thing where you load up 6 images into comfy and through some...thing it uses them similar to a reference. I see you're using mostly the same style, so a single lora would suffice for that imo.

Anonymous 07/06/2025 (Sun) 06:14:54 No. 29139

>>28839 UOOOOOO ANRIETTA-SAMAAAA 😭 >>28951 Cute! That squid character reminds me of the H.O.U.W.A.R. hat from TF2

Anonymous 07/06/2025 (Sun) 22:55:57 No. 29401

[ Return / Catalog / Top ]

[ Manage Board / Moderate Thread

Delete

Password Delete only files (Removes the file reference to the posts) Delete media (Removes the saved files from the server)

Report

Reason Global

no cookies?