/mg/ - AI Monster Girls Thread

/mg/ - Monster Girls

Main monster girl board

Mode: Reply

Name
Options
Subject
Message	Max message length: 20000
files	Drag files here to upload or click here to select them 0.00 / 50.00 MB Max file size: 30.00 MB Max files: 5 Supported file types: GIF, JPG, PNG, WebM, OGG, and more

E-mail
Password	(used to delete files and posts)
Misc

Remember to follow the Rules

New front-end has been installed, if your page layout became messed up you need to clear your browser cache. Post in >>>/site/2 if you are having issues.

Remember to bookmark emergency backup pages in case the site goes down: https://rentry.org/capybackup and the mirror https://rentry.co/capybackup

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 07/15/2025 (Tue) 05:49:27 No. 31121

Requesting a Lamia wearing a sombrero and a poncho. Bonus points if she has a comically large fake mustache attached.

Anonymous 07/15/2025 (Tue) 16:16:09 No. 31179

>>31121 >a lamia How about 240+ lamias? Man, this was fun. Also learned a lot: nnaillousXlSeries_n40naillousv2Xl (N40 for short) knows a ton of artists by default. Not adding any artist tags usually resulted in trash (near ms paint levels, look at first gens in the collage). Adding "masterpiece, best quality" as first 2 tags makes the end result a little better, but using artist tags, or something akin to them seems like a much better choice. Changing tags and queuing generations faster than your gpu can do them seems like the optimal way to use the tool (I've yet to get this month's electricity bill though, haha). After you get a ton you just sort through them, upscale/edit the good ones. Its like photography- first you spam the shutter, then you pick the best. If you want I can upscale some pics from this set (ie raise quality, fix fucky faces/hands/details), just show which one in the collage and I'll try.

Anonymous 07/15/2025 (Tue) 16:25:27 No. 31183

>>31179 Some more highlights.

Anonymous 07/15/2025 (Tue) 20:43:15 No. 31231

>>31121 Based request. I thought it would be apropo to make her a rattlesnake, and i like the other anon's idea of making her shred on a guitar.

Anonymous 07/15/2025 (Tue) 23:03:55 No. 31256

>>31231 >Based That's not how you use this word. >rattlesnake Good idea though.

Anonymous 07/16/2025 (Wed) 01:45:32 No. 31297

>>31179 >>31183 >>31231 Very good, thanks hermanos. I'm curious about that sequence mixing a bunch of different anime characters or distinct artstyles with the request, did those come about naturally or did you put them there yourself? >Rattlesnake I actually considered suggesting that myself, but I didn't think there'd be enough for the model to pick it up.

Anonymous 07/16/2025 (Wed) 03:52:18 No. 31324

man four arms really does offer up a lot of unique posing potential.

Anonymous 07/16/2025 (Wed) 09:44:10 No. 31361

>>31297 I mentioned in the first post I was experimenting with artist tags. The only ones I explicitly prompted were Tomoko, Konata, Ranni and "blonde_hair, long_hair". Stocking came out when trying to force it to use the PSG artstyle. Asuka/Nadia came out when using the OptionalTypo artist tag (the guy draws them a lot, and has a gainax-ish style in general), the FLCL bitch came out when using sadamoto_yoshiyuki (official artist for many core gainax things). I also tried making one in the style of van gogh's starry night, but the model thought I wanted fate's van gogh character. For most generations I didin't put any details on the character so the model "decided" itself what hair colours, styles, faces, tail colours, etc to use.

Anonymous 07/16/2025 (Wed) 09:47:56 No. 31362

>>31361 >>31297 >>31179 Also worth noting is that that collage has everything I generated then, in chronological order starting from top left. So you might also see when I added new tags or refined the idea (sky, sun, dancing, perspective, dynamic, 2 girls, etc).

Anonymous 07/16/2025 (Wed) 10:03:21 No. 31363

>>31324 Its like having 2 characters posing at once. I wonder if something similar could be achieved with 4 eyes...

Anonymous 07/16/2025 (Wed) 12:54:48 No. 31382

https://github.com/lllyasviel/FramePack?tab=readme-ov-file https://github.com/lllyasviel/FramePack/discussions/53 Finally managed to get it running on my pc with 5060 ti16gb. It didn't work initially, had to also follow the guide in the second link for the 50 series cards, then (eventually) it finally worked. The results are definitely slop-ish, but the potential is there! The 10 sec one took around 50 minutes, the shorter one around 20 (I think, didn't measure it unfortunately). lllyasviel says his 4090 does around 1 frame in 1.5-2.5 seconds. So relative performance appears to be more or less alright, but I'd have to measure it better. Either way, just check your GPU relative to his and mine and you should get a rough idea what generation times you can expect. Also of note, all my 32gb of RAM were filled during generation, but not all VRAM according to task manager. Websites were definitely deleted from ram because after clicking on a few tabs they had to be reloaded. Do what you will with this information. Reading into it further (https://github.com/lllyasviel/FramePack/issues/151#issuecomment-2817054649) seems like RAM speeds and SSD/HDD speeds also have non negligible effects. Hey 5090 anon, I have a request, try running it on your machine and post results and generation times, pretty please? Do whatever you want, I only want to know how long it takes on the highest end gpu.

Anonymous 07/17/2025 (Thu) 18:52:48 No. 31647

>>30739 >Need more amazonian Gazers. <huh, lets just check if N40 can do Gazers without loras, nothing much <201 pics later I can't wait for the electricity bill! Anyway, experimented again, this time with very low step counts to save time (and also electricity, lul). Turns out you can go VERY low, at just 10 steps you start getting some ok stuff. Going above 20 seems pointless from my testing, 15 would be the fast option for spamming images, then just upscale the good ones (which I still didn't do, heh...). Aside for that, tested artists again. I find it funny how vanripper makes them more clothed.

Anonymous 07/22/2025 (Tue) 21:44:47 No. 33064

Anonymous 07/25/2025 (Fri) 03:23:22 No. 33603

Anonymous 07/26/2025 (Sat) 21:36:36 No. 34304

Anonymous 07/29/2025 (Tue) 03:18:08 No. 35121

Anonymous 07/30/2025 (Wed) 21:37:52 No. 35549

Anonymous 07/31/2025 (Thu) 06:55:03 No. 35622

>>35598 >>35599 Great stuff. Catgirl wrapping her tail around dick is peak

Anonymous 08/09/2025 (Sat) 12:57:57 No. 37974

I've been trying to do imagie2videos going with comfyui but holy hell it takes forever and the results are shit. Anyone got any good ideas when it comes to img2vid?

Anonymous 08/09/2025 (Sat) 20:23:24 No. 38089

Anonymous 08/09/2025 (Sat) 21:13:05 No. 38109

Requesting Ilias drawn in KC's art style.

Anonymous 08/10/2025 (Sun) 00:20:07 No. 38134

>>37974 Framepack by illyasviel. Also takes forever, but sometimes the results are nice.

Anonymous 08/11/2025 (Mon) 05:47:14 No. 38427

>>31382 Sorry I've been busy a lot lately. I tried to get it working a while ago after your request, but my results were twice as slow as Illyasviel's. I looked into it and thanks to you, I discovered there's a problem with my build. Basically, my PCI-E slot is limited to 2 lanes, or 1/8th of the max memory bandwidth. No idea what the fuck has happened. Seems it's likely either my motherboard, cpu socket pins, or the GPU. I'll have to look at it later. So, thanks for that. It's good to know about. Now that I've seen it, I can't unsee it, seems I'll have to do something painful to fix my pc soon... But because of that, I can't provide you with a reliable benchmark at this time. I never noticed because memory bandwidth isn't too much of a bottleneck on FPS, especially at PCi-E 5.0. >>38134 FramePack Studio (fork of framepack) gave me slightly better times and has more options, for things like loras or image loops. >>38109 I'll give it a try soon.

Anonymous 08/11/2025 (Mon) 08:48:00 No. 38465

>>38109 >>38427 Gave it a try, did some fun ones, and some lewd ones since there wasn't a specific thing requested. Tell me if you want any of the individual images. Just upload a crop of the grid ones you want, or I guess give me numbers by row and column and some description. The NSFW ones are a bit sloppy because I used a bunch of wildcard files I have together to generate them, and there's probably some incompatible tag overlap. I can make a lot of better stuff deliberately, so feel free to ask if you have a specific position or thing you want.

Anonymous 08/11/2025 (Mon) 08:50:07 No. 38466

>>38465

Anonymous 08/11/2025 (Mon) 08:51:04 No. 38468

>>38466

Anonymous 08/11/2025 (Mon) 08:55:33 No. 38470

>>38468 Actually, scratch everything I said about uploading the originals. I thought the grid was preview, but it turns out it embeds the full images in all of them. So I had to upload it in chunks like this. This one image is downscaled to fit though, so you can ask for a bigger original image for this one specifically.

Anonymous 08/13/2025 (Wed) 22:54:21 No. 38932

Anonymous 08/14/2025 (Thu) 08:58:55 No. 38995

>>38465 >>38466 >>38468 >>38470 Thanks anon, but I gotta be honest, it doesn't feel like it can quite replicate it yet, it's still missing something. Also lmao on those wormholes in the first NSFW pic.

Anonymous 08/17/2025 (Sun) 16:50:38 No. 39612

>Switch from pony to Illustrator as checkpoint >Results are 10X better Well okay then.

Anonymous 08/19/2025 (Tue) 02:58:22 No. 39784

>>39612 Aw dude, if you've been using pony for that long I feel real sorry for you. It's a night and day difference.

Anonymous 08/21/2025 (Thu) 19:27:40 No. 40478

It's kinda interesting how good AI has gotten with tentacles, it used to fuck them up really badly.

Anonymous 08/21/2025 (Thu) 19:41:20 No. 40489

>>39784 It's really fucking weird learning I've basically been doing this with one hand tied behind my back. Same with moving from automatic111 to comfy ui. Once you get it, it's so, so much better

Anonymous 09/01/2025 (Mon) 02:03:51 No. 42846

>>40478 AI is 100% the future. The rate at which it's improved in just the past two years is astonishing. I was a skeptic back then, but I'm convinced now. Really, AI art generation is sort of outpacing AI chatbots. I bet that by the end of the decade, artists will have become completely obsolete. It's so cheap and easy to generate new art now, and the speed at which it's done. I can only imagine with 5 more years of improvement what it will be like. It's gonna be so nice too. No need to embarrass yourself by asking some random artist to draw some niche character doing some lewd or obscene act, pay an exorbitant amount of money, and wait 2-6 months for the final piece. Now you can just put in a text prompt, wait a couple minutes, and boom, it's done. And for less than a dollar.

Anonymous 09/01/2025 (Mon) 07:04:44 No. 42866

Anonymous 09/01/2025 (Mon) 07:10:57 No. 42867

>>40478 >>42846 yes, yes it is very good with them now.

Anonymous 09/01/2025 (Mon) 09:36:56 No. 42890

>>42866 >it even got the tattoo pattern right Neat

Anonymous 09/01/2025 (Mon) 14:18:57 No. 42944

>>42890

Anonymous 09/02/2025 (Tue) 23:56:14 No. 43232

Anonymous 09/03/2025 (Wed) 18:13:53 No. 43415

>>42846 Ehhh, yes and no. Pin up art of established characters? Yeah, 100%. But fanart in general? Ehhh. It still struggles a lot with multiple distinct characters. trying to make two characters is hard enough, 3+ and you're just asking for trouble. If your character is too niche it won't have any art to make a good lora from. and if there's less artists there's less people to fix that issue. Also making OCs will still require an artist if you really wanna get it right. We might see the end of commission artists for sure tho.

Anonymous 09/03/2025 (Wed) 19:53:35 No. 43440

Alps deserve love too

Anonymous 09/03/2025 (Wed) 21:48:48 No. 43457

Anonymous 09/04/2025 (Thu) 17:18:07 No. 43596

>>43440 What model are you using that's capable of (somewhat) legible text?

Anonymous 09/04/2025 (Thu) 17:25:04 No. 43597

>>43596 https://civitai.com/models/827184/wai-nsfw-illustrious-sdxl It works decent enough for text, though it's still takes a few tries to get it the way you want it

Anonymous 09/04/2025 (Thu) 17:26:11 No. 43598

>>43596 >>43597 Should not I'm using v14 and haven't tried v15 yet

Anonymous 09/04/2025 (Thu) 17:26:47 No. 43599

>>43598 >Should not should note*

Anonymous 09/04/2025 (Thu) 17:55:41 No. 43601

Anonymous 09/04/2025 (Thu) 23:35:38 No. 43649

>>43597 Ironically the model I'm already using and recommended in the guide. I just thought it was completely incapable of text since that's what people say about anything from SDXL, (Illustrious, NoobAI, Pony, etc.) Thanks for broadening my horizons to how great the model is lol

Anonymous 09/05/2025 (Fri) 00:33:38 No. 43651