BBCode
(for deletion)
  • Allowed file types are: gif, jpg, jpeg, png, webp, webm, mp4
  • Maximum file size allowed is 25000 KB.
  • Images greater than 255 * 255 pixels will be thumbnailed.
  • 110 unique users in the last 30 minutes (including lurkers)




    There's this new image model named Anima and it's pretty good for making anime guro. Here's Toga with her legs chopped off. I wanted to make the bones stick out, but i don't know how. If anyone has any suggestions on how to do it, you're welcome to contribute here.
    1 posts omitted. Click Reply to view.
    >>
    >>18151
    >>18152
    Almost forgot to mention that these were enhanced by NoobAI.
    >>
    It can make decapitated heads. Still want to make both a headless body and its decapitated head
    >>
    Amazing! Do you take requests?
    >>
    >>18153
    is NoobAI better then Illustrious at guro?
    >>
    >>18152
    nice stiches

    Starting a thread to share my testing and findings with the new Anima model, download here: https://huggingface.co/circlestone-labs/Anima

    My main goal and fetish is cooking and cannibalism so thats where my testing's focused on. Will be posting generated images.

    Details to know and why I'm interested:
    -Its a new model in testing that generates very fast without much hastle, expecting the full release to be slower but people already made checkpoints to speed it up.
    -Model advertises that can do both tags and natural language or a mix of both. With clever wording, you can get good results.

    Issues:
    -Its a model on preview and its not the full thing yet
    -Its a new architecture and its mpt compatible with flux or sdxl/illustrious loras, requires its own lora training.
    -Proper guro things like beheading, guts, blood, etc etc aren't implemented (or I haven't found the wording for them)
    -Doesn't work with controlnet, doesn't work with regional prompting.

    Comment too long, view post No.16940 to see the full comment.
    38 posts omitted. Click Reply to view.
    >>
    Hey, been a bit. I've been inspired by Mrik's checkpoint and loras so I've been working on a multiconcept lora for my specific fetishes. The lora includes:
    -Spitroast
    -Cooked bodies
    -Cooked breasts
    -Grilled penises
    -Seasoned bodies
    -Carrot in anus/pussy

    Its not finished and I might be doing things wrong. If someone has has experience on them I'd love to read about it. A preview on the cooked breasts.
    >>
    >>18972
    Cooked breasts are ana amzing fetish


    I would reccoemnd doing separate LoRAs for each of those concepts. It usually works better this way
    >>
    >>19015
    Thanks. I'll try doing separate loras in the future. Meanwhile I work on the loras, I'll post the current "big lora" I made so people can use it in the mean time. It is not perfect, it breaks a bit, but it should produce results.

    Download link:
    https://www.mediafire.com/file/f2d6xlt2wpu1yru/CanniMix.safetensors/file

    The general trigger word is "cannibalism" and more specific ones are:
    -"cooked_body" For cooked people, may need "dark skin" if the cooking isn't happening.
    -"cooked_breast" For cooked breasts, as before may need to specify "dark skin breasts".
    -"spitroasted" "impaled" "all the way through" for spitroasts. I add "steel skewer" and similars to specify.
    -"penis on grill" and "grilling penis" for dicks that go on grills. Doesn't work that well but it provides cooked dicks. May need to specify.
    -"dumbbune" an artist I really enjoy and I've tried to get the style from.
    -"seasoned" for seasoned bodies. May need to use "spices" or "sauce" for extras.

    Comment too long, view post No.19027 to see the full comment.
    >>
    >>19027
    terrific
    >>
    The people behind kohya_ss have started working on providing anima with controlnet. I'm seeing some decent results, but of course its still in the works and the base model isn't even fully released. Links:
    https://github.com/kohya-ss/ComfyUI-Anima-LLLite
    https://huggingface.co/kohya-ss/Anima-LLLite

    https://files.catbox.moe/nxkui9.mp4

    or

    https://file.garden/aecQ3YtN8glaOMdJ/frame-crux.mp4

    I'm happy to have finally done the full concept I had for this video, after having set the project aside for a while... but the quality is very meh.

    Further, Wan 2.7 clearly has a limited understanding of how chainsaws operate and where cuts happen relative to where the blade of the chainsaw is placed. I've seen some chainsaw confusion in the Grok videos too, but at least they generally have much higher video quality than I ended up with.

    It's possible if I did this over again at 1080p instead of 720p I wouldn't get so much generational loss from using the last frame of one 15-second segment as the starter frame for the next. The main thing that stopped me from using 1080p is that I was being cheap, knowing (as turned out to be true) I'd end up generating a few dozens video clips just to get the final four clips that were on-target enough to be usable.

    If anyone has any hints that would help make Wan 2.7 either smarter about chainsaws, or better with video quality when stringing multiple segments together, I'd love to get some pointers.
    >>
    love it carry on being 'cheap'

    Here it is: https://files.catbox.moe/6gf0op.mp4

    I posted a rough cut of what I was trying to do with this video weeks ago, with gaps in the audio, awful voice-acting in some places where there was audio, and ending with only one of two breasts removed because Wan 2.2 just couldn't figure that out -- it kept wanting to re-cut the already missing breast instead.

    This thread, however: https://boards.guro.cx/ai/koko.php?res=19033&page=0

    ...inspired me to give a mostly-uncensored version of Wan 2.7 on Venice.ai a try. With a fair amount of trial and error, Venice.ai and Wan 2.7 finally produced something close enough to what I wanted for the breast cutting itself for me to use in a final cut of the full video. I also used the new Wan 2.7 models to enhance some of the older video and add or redo audio. Some old bits of Wan 2.2 stuff remain in the final cut.

    I've decided to take the approach for now of not worrying too much about little glitches and inconsistencies and less-than-perfect scene transitions, in favor of simply moving on to new projects and trying to use what I've learned going forward.
    4 posts omitted. Click Reply to view.
    >>
    >>19208
    That did it. It's just the quirks of this board. For those who aren't seeing it, and only getting the attached if you directly click the link, instead right click on either of the links galathien posted, and then choose "Save Link As". I'm using Firefox, so whatever your equivalent would be. Then it will download to your hard drive..

    Okay... With that out of the way... HOLY FUCK BALLS, that was HOT!!! Jesus F-ing Christ. Love how she took her time, love cunt all secured to the wall with no tits left, and LOVED her expression when said she still has a "Lovely bald Cunny". LOVED IT!
    >>
    Also, don't know if it's what you were going for, but the SMOKING HOT cunt who gets her tits cut off is almost a dead ringer for big titttied Milana Vayntrub. The one who used to do the AT&T adds.

    If that was what you were going for, UPROARIOUS APPLAUSE. You succeeded with flying colors! If it wasn't... Same thing!
    >>
    >>19210
    > If that was what you were going for...

    Just a happy coincidence... and yes, Milana Vayntrub is hot to think about in that scene. Yum.
    >>
    Now this is very nice, it's like a full-on scene... She's enjoying slicing up her perky titties nice and slow.
    >>
    >>19242
    > Now this is very nice, it's like a full-on scene...

    Thanks. And that's what I really want to work on most, rather than just a few seconds of "See this thing happen. OK, done".

    Not that I won't occasionally indulge in short clips, but want to keep pushing my skill set for making longer pieces.

    All images depict women 18 years of age or older.
    4 posts omitted. Click Reply to view.
    >>
    >>
    >>
    >>
    >>

    Kinda Missing the Angel corps gals.
    50 posts omitted. Click Reply to view.
    >>
    i think i generated like 500 images just to get the ones I liked here. I gotta get better rather than just rolling the digital dice
    >>
    >>18934
    Damn you did this brute force, that's actually the opposite kind of impressive I thought it was, but no less so!

    Excellent work~! Thank you for sharing~
    >>
    That doesnt invclude time in comic creation, writing and post effects
    >>
    >>18945
    What programs/workflows do you use for manual work? I'm mainly focused on Krita for drawing and painting, I've been thinking about getting into doing comics but the amount of work, as you have confirmed is daunting and I am not sure about what best practices are for getting good results from that kind of medium.
    >>
    I used clip studio pro for both making the comic and for editing of the images

    Ashley was preyed upon and digested.
    19 posts omitted. Click Reply to view.
    >>
    >>
    >>
    File: 1000094770.mp4
    (6.21 MB, 1080x2160)
    6515417
    >>17549
    Another angle version.

    >>
    >>
    Woah this is brutal, and I love it.
    >>
    >>
    That's really nice. How is it generated?
    >>

    ANONS,

    A lot of us here create AI art but have zero motivation lately. We have the tools and talent, we just need a push.

    Let’s start a monthly competition.

    How it works:
    - One theme per month (first one posted below, then we vote on future themes)
    - Submit your best piece(s) for that theme
    - End of month we vote and pick a winner

    Prize for the first month: 10,000 Civitai Buzz from me.


    Comment too long, view post No.18253 to see the full comment.
    5 posts omitted. Click Reply to view.
    >>
    >>18326
    That is better then I saw grok doin in the past.

    I assume you went directly txt2vid without doing a txt2img first?
    >>
    >>
    >>18330
    text to image then image to video, all in grok. >>18313
    >>
    >- AI only (any model/tool)

    Well I was intrigued for like 5 secs.
    >>
    In general what do people think about this kind of monthly "competition"/best of the month media?

    does it motivate you? Should the board be left to go as it was going before?

    Hello. Any DBD fan or Meg Thomas fan over here? I don´t even know how to use these loras and programs people share here. I use Motion Muse, which is not optimized for guro, and play with the prompts. Hope u like them
    1 posts omitted. Click Reply to view.
    >>
    >>
    >>
    >>
    what prompts do you use for the head fuck?
    >>
    >>18561
    First of all, I have to get the image before transform it into video. I.e: The figure is resting on a long fat erect penis of a man seated beneath it. The penis is sinking under the figure, the tip of which is invisible because it is hidden within. Man holds figure's brain by both sides with his hands. Slightly zoom out on camera. Or "
    Girl face expression remains exactly the same. whole Bloody brain is exposed at the level of her eyebrows, . From the upper rear appears the figure of a man, of whom only the lower and flexed half is visible. His erect male reproductive organ is inserted deep into the girl's brain. Tip of the dick is invisible because its inserted inside brain craving a hole on brains surface.

    After I get the edited image, there are 2 models where the AI can fuck holes without getting bugged AFAIK. They are "time stop sex" and "Love Doll Sex", both done by an user called Hattori.

    or "How I lost my virginity"
    119 posts omitted. Click Reply to view.
    >>
    >>
    >>
    >>
    >>
    >>18736
    She's OK. That's just a strap-on.

    Hi I'm GM or Blackheart, been a guro fan for 20 years, but never much developed my talent to a point where I was happy with what I was making, but after nearly 2 years of learning how to use image gen and the resources of this helpful community I'm beginning to make guro pieces I am proud of. Today I share one with this community as way of saying thanks for all the years of making this pervert feel a little less alone in the world.

    Made in ComfyUI using Default V6 PonyXL a few LoRAs from around this board and lots of GIMP.
    65 posts omitted. Click Reply to view.
    >>
    Decided it was good enough to polish up, didn't need much after upscaling and some img2img.
    >>
    >>
    >>
    Made in ComfyUI Using:
    AnimaYume v4.0 + Upscaled with Flux 2 Klein.
    >>

    Hello everyone,

    I'm knew and was able to find good hanging LoRAs. But I still need help finding a good one for decapitation, guillotines, etc. Any recommendations?

    ps: Imagine is not AI generated
    >>
    bhdsx, rbqinori, gcheckpoint
    >>
    >>17894
    Where we can find this loras?
    >>
    >>18465
    These Loras can be found on this website. By the way, where can I find the hanging Loras you mentioned?
    >>
    >>18519
    https://e-hentai.org/g/2928445/7da13650d4/

    i only know about those, they are for pony, but works with illust too, not the best, for hanging i would've used either rbqinori or gcheckpoint

    I usually don't have a reason to do stuff like this, but everytime I wanna try and generate pantypoop/soiling stuff, one of the things I go to is to generate a generic girl that does the pantypooping.

    My design for her is that she had medium-size breasts, brown hair and eyes, and a beanie, specifically orange, but my dumbass and horny brain decided to make a blue beanie girl, and now im stuck here wondering what to do naming them instead of referring to them by the color of their beanies

    Like if I wanna refer to one of them, I call them sumn like "orange beanie" and it feels cumbersome and awkward to type. So here they are. What should I call them according to yall?

    After giving up on Grok, I found this option. While this result could use some refining, it's just my second try with Wan 2.7 and this starter image.

    Full video result: https://files.catbox.moe/db0ob7.mp4

    Prompt:

    > A terrified redhead has been brutally nailed down to this wooden frame, from which she cannot escape. She is securely attached at her wrists and ankles.

    Woman with chainsaw uses it to brutally cut off the redhead's legs, at the upper thigh, one leg at a time. Blood spatters as the saw cuts through flesh and bone. It only takes two seconds to cut through each leg.

    Since each leg is nailed firmly to the wooden frame, as each leg falls away it pivots at the ankle as it drops to the ground.

    The victim is screaming in agony, while the woman with the chainsaw is enjoying herself immensely, becoming lightly spattered with blood.

    Comment too long, view post No.17934 to see the full comment.
    5 posts omitted. Click Reply to view.
    >>
    Does Vanice.ai support first-last-frame and continuous generation?
    >>
    >>17955
    I haven't explored all the option yet, but there was definitely a first/last frame Wan 2.7 option, and also Wan 2.7 Reference, which can use multiple reference images.
    >>
    how is venice compared to a pay per gen site like dezgo
    >>
    >>17963
    > how is venice compared to a pay per gen site like dezgo

    I certainly don't know dezgo specifically, and I never spend that much time at any site that parcels out "gems" for processing credits.
    >>
    How does it do with hanging

    Delete post: []
    [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35] [36]