BellyPunch PonyXL lora
Triggers: stomach_punch | score_9,source_anime
Weight: 1.0
Recommended Epoch: Read the ReadMe file. But basically low epoch = clothed, high epoch = unclothed

Download: https://mega.nz/folder/3io1RZIB#9CkkuRnivAX3zNKm15gjuw


----------------
I also made an SD1.5 version of this lora which I posed here a while ago. I've been getting some google drive requests since I moved the files to mega so I'll repost the link here

Trigger: harapan (Supplementary: hand on stomach, (1boy:1.5))
Download (SD1.5): https://mega.nz/folder/m3pU0QDT#_AMdp89kKVPI6wAH5WcESw

Impaled PonyXL lora

Triggers: impaled | score_9,source_anime
Supplementary: guro, blood
Weight: 1.0

Download: https://mega.nz/folder/TngFiDjA#x9yxe21YSP00KPe72OqfYQ

Haven't done nearly enough testing on this. Made it on a whim, and don't plan to thoroughly test this out anytime soon. Can't really offer any advice. Figured I would share it anyways rather than sit on this since the amount of guro XL lora's are pretty low.

>>687
a lora for hanging would be nice

>>688
There already is one which is why I didn't make one.

https://rentry.org/R-18G_Stable_Diffusion#xl-hang-board-test

>>689
how about broken bones? as far as i know, there are quite some loras about amputee, but none about broken bones.

>>689
the person who made that hanging lora (https://www.pixiv.net/users/86218912)also made a impale lora.
your impalement lora fits charactor loras much better than his except too much amputation is included. if you train a hanging lora maybe yours could fit charactor lora much better

>>691
Alright I'll give it a shot then. I was under the impression that it worked generally well so there was no point to make the same thing.
>>690
Seems interesting I'll try that out too.

Here from /h/, just wanted to say thanks for the LORA

>>692
>>691
>>688
Hanging PonyXL lora v2
Author: Entropy (https://www.pixiv.net/en/users/32339993)

Triggers: entropy_hang
Supplementary: rope on neck, rope around neck, corpse, arms_behind_back
Weight: 1.0 ~ 1.8
Reccomended Epoch: 20

DOWNLOAD: https://mega.nz/folder/DzQRyCbY#RUwVu3RkK4CyaSU3-AdM0g

Notes: Characters often have their feet touching the ground or is on tiptoes. (When using full body)
Without full body general "camera" height is low enough to imply the character has their feet touching the ground. RNG may sometimes be in your favour and generate an actual "hanging" image.

Works very well with character prompts, style lora's and concept lora's. (Haven't tried character lora's but should work). Used a unique keyword in this case instead of a prexisting keyword. As https://rentry.org/R-18G_Stable_Diffusion#xl-hang-board-test mentioned, using keywords such as hanging, hanged will generate bondage hanging instead. (Reason why such keywords wern't used as the trigger word for lora)
Also trained to do side_view (this works), gens. Also attempted to do back_view (from behind) but it seemed to fail (I think too low of a sample size).

Megumin Character Prompt - Quadruple amputation Lora - Side view (Attached image)

With character prompts I found a higher weight was required (1.5~1.8). Basic prompts would generate a "hang" with 1.0 but I'm guessing the information is being overwritten by the character or detailed outfits.

I also found mixing this Lora & the other Hanging lora generated some decent results. (All XYZ graphs in mega folder)


>>690
Currently trying to make one but I'm struggling to get enough good images for it. Some help in gathering training images would be appreciated. I guess thats why there isn't a lora for it though.

>>693
You're welcome. Glad someone else is using it & enjoying it.

>>694
thats good hanging lora. maybe its time to get a corpse lora that represent not so fresh corpses, maybe based on zombies?

>>694
your hanging lora fits charactor loras much better than his, good job!
he also made headshot and neck snap, and judging from how impalement and hanging lora worked, you could make a version that fits charactor loras much better

>>687
did some testing
Got a few nice ones but many misses
thank you

QuadrupleAmputee (Flesh) PonyXL lora v2
Author: Entropy (https://www.pixiv.net/en/users/32339993)

Triggers: quadruple_amputee
Supplementary: guro, blood (Adds blood to missing limb, not using guro/blood will result in clean but fleshy wound)
Weight: 1.0-1.3
Reccomended Epoch: 40

Download: https://mega.nz/folder/yyA1jYIa#m5KjCXGoj3F7j64XDmlBGQ
Please Read the ReadMe and or view the included XYZ graphs for more information.

Made this Lora for a recent mini project/manga I just finished. https://www.pixiv.net/en/artworks/118347437
Still translating it to english so only japanese is available but I'd appreciate if anyone interested in this stuff would check it out.

>>695
Working on a corpse lora now. Seems doable hopefully I will have something done in a few days

>>689
Hey, kudos for sharing that hanging Lora - best one I have found for Pony yet. Works absolutly great!

>>685
Experimenting with lore and realistic PONY models.

>>700
The parameters are saved, just download the image and place it in automatic1111.

Models:
https://civitai.com/models/400801/pianomix
https://civitai.com/models/393905?modelVersionId=475501
https://civitai.com/models/409856?modelVersionId=484151
https://civitai.com/models/330790?modelVersionId=463782

Faces were corrected with help adetaler: https://github.com/Bing-su/adetailer

>>700
>>701
>>702
>>703

Sorry, photo-realistic AI images aren't allowed on GuroChan.

Thanks!

Corpse PonyXL lora
Author: Entropy (https://www.pixiv.net/en/users/32339993)

Triggers: corpse
Supplementary: guro, blood, empty eyes, pale_skin, grey_skin,
Weight: 1.0-1.5
Reccomended Epoch: 40

Download: https://mega.nz/folder/WmxxiBZK#C_APm3H4nDXEuETHguiyjA

Notes: Creates a desaturated skin tone as well as a limp face to simulate an old corpse (Not to the point of rotting).
lying on back / on_back -> works
lying on stomach -> seems to cause problems just use lying instead and let it gacha (something like breast_press may help your odds)

To create a strong effect modify both the trigger word and the lora weight. I find 1.2 for both good at creating a desired effect. Raising just one to a higher value >1.5 will create a very dead looking corpse but high trigger weight causes blood to form and high lora weight well just looks weird (see XYZ graph).

Consider adding "rating_safe" if you want a clothed corpse. A large percentage of the training data was nude corpses so it may create nude corpses even if you did not specify such elements. Although it may still sometimes generate genitalia, in which case you should add clothing keywords that cover that body part. (ex. panties, bra etc.)

I'm really not into necrophilia but unfortunately (or fortunately for some of you) a good percentage of the training data was exactly that...
so it's pretty good at generating that too... do whatever you want with that information

Struggles with character prompts & lora's. Simply send your gen to img2img remove any character trigger words (keep all other prompts like clothing, pose etc.) and let the corpse lora run without the character prompts interference. (See example 1-1 & 1-2)
Example 1-1 was generated with the character prompt "asuna \(blue archive\), asuna \(bunny\) \(blue archive\)" This seems to have significantly obstructed the corpse lora from creating its effect. Skin appears normal but slight facial expressions have been applied. Example 1-2 shows the result after using img2img (I used 0.5 denoise but I doubt it really matters, figure out whats best for you) and REMOVING the above character prompt. No other prompt data was modified. I suggest going back with inpaint to fix clothing details that were lost due to the character prompt not being there, if any.

Works with other Lora's. See attached image-> working lora mixing with my previous hanging lora.

>>695
Thanks for the idea I'm actually really liking this Lora. I tried using zombies but I ran into the issue of most good zombie images having stiches so I had to abandon using them outside of a few that I did find that would work.

I've also created a AI generated guro manga using a few of the Lora's I've posted as well as a few more from the community. I would appreciate it if you would check it out (Just released the English version today)

pic related is the manga. 31 pages.

https://www.pixiv.net/en/artworks/118512043

https://civitai.com/models/446867/hunting-trophy?modelVersionId=497626

I discovered this LORA on civitAI. It is unexpectedly very practical. It can be said that I have found a treasure.

Strength:1

Required prompt words:(hunting trophy:0.7),(shield:0.8),|wall,

HOW:(wholebody:1.2),|halfbody,face,breast,|head,|head,feet,soles,|necklace,pussy,(strength1.5)|(head:1.2),butt,through wall,stuck,stationary restraints,|

los:looking afar,|looking at viewer,|looking up,|looking down,

view:FrontView,|TopView,|SideViewL,|SideViewR,|BottomView,

>>705
maybe you should retrain the impale lora, with amputee prompt added in database. right now too much amputation is included, and there is no negetive prompt to control that.

>>707
yeah, this one is amazing.

>>706
i love the long story format. The empty eyes should happen at a much later point of the story imho.


PS we NEED an eye fucking LoRA!!!

>>967
yeah that's definitely on my list of Lora's to make. I ended up just using inpaint to make that one scene in the story.

Can I suggest two more LoRAs to make?

1. Is girl tied to a sandbag and used as a punching bag.

I collected a dataset, sending it to your pixiv. But you should probably take a look and throw away half of images.

2. Is a girl tied to shield. Maybe turned to stone, and maybe filled with arrows.

Punching Bag PonyXL lora
Author: Entropy (https://www.pixiv.net/en/users/32339993)

Triggers: punching_bag
Supplementary: restrained, tied, injury
Weight: 1.0
Reccomended Epoch: 40

Download: https://mega.nz/folder/qjRUTB7b#TqnW2XDF94aHURZU3RfMsA

Notes: Creates an image of a girl tied to a punching bag. The supplementary tags Tied / restrained will result in the girl being tied to the punching bag. Without those tags the result will be the girl just being in front of the bag. Arms and Legs being tied behind the punching bag are unfortunately a bit of a gacha generation. Keep trigger word weight at 1.0, higher weights will result in a fried image, if you are encountering collisions between Loras increase lora weight instead. Mixes decently well with other Loras, styles and concepts. Another anon has tried using them with realistic style and this lora seems to work with them. Image with girl being punched uses two lora's this one and a belly punch lora. Unfortunately both are a bit gacha so it may be easier to just make the punching_bar image than use img2img to add the bellypunch but it does work with txt2img (just required a lot of rolls)

>>1397
Thanks for the images. As a side note if any anon's want to request a Lora if you send me some training images (atleast 20 but more is better) it will really help me and make it so I can pop Lora's out real quick.

>>1512
Used both bellypunch and punching bag for this one

>>1514
>>1512
The punching bag is awesome!
Thank you for making it.

If you ever make a v2 please contact me, I have plenety of punching bag generated with this version

sometime you punch too hard and she breaks

>>1518
or punched in the breast

Meat Shield PonyXL lora
Author: Entropy (https://www.pixiv.net/en/users/32339993)

Triggers: meat_shield
Supplementary: restrained, bondage, bdsm, 1boy, holding shield,
Weight: 1.0
Reccomended Epoch: 20

Download: https://mega.nz/folder/bjJV0IZD#QQhDRABTZtRHnJh-Q1znqw

Notes: Creates an image of a girl attached to a shield. 1boy & holding shield will result in a man behding the shield holding it. Has a preference top create images with crotch exposed. Consider using prompts to cover it up like panties or something if you want that part hidden. rating_safe may also help. Same applied to breasts.

>>1398
Whoops made this a few days ago and I completely forgot to post it.

>>687
so far I got the best results with the safetensor with 10 epochs.

What prompt do you use to have the pole emerging from her mouth?

>>1569
I used epoch4 for the image but to be honest I had to roll over and over to get that. It's extremely inconsistent, gacha Lora moment.

https://files.catbox.moe/59pxa1.png is the image in >>687
You can copy the prompt from it if that helps but it's going to be very gacha.

I recommend doing txt2txt then sending a gen you like (without mouth spike) into paint/photoshop and draw a pole from the mouth (can be shit drawing, literally just draw a solid brown/red or whatever color you want line from the mouth). Then sending that into inpaint. It should recognize that its supposed to be a pole coming out of the mouth. pic related. I generated one image then slapped a pole in the mouth and send it to inpaint. Takes an extra few mins but it's so easy that it's worth it. Instead of waiting for the perfect seed, I think this is a much better use of your time. Sorry I don't have a txt2txt solution. Hope this helps

Arrow in Body PonyXL lora
Author: Entropy (https://www.pixiv.net/en/users/32339993)

Triggers: arrow_in_body, arrow_\(projectile\)
Supplementary: injury, blood, guro, and other blood related triggers
Weight: 1.0+
Reccomended Epoch: 30

Download: https://mega.nz/folder/2rZVmCSD#9M3QJrgoFYe06Xqtro_w8g

Notes: Creates an image of a person (only trained on 1girl but should? translate to male characters) shot by an arrow. Use both "arrow_in_body" and "arrow_\(projectile\)" for triggers. Using only one will still work but results may not be as clean. Increase Lora weight for more arrows, I have tested upto 2.0 weight which still appears clean (Higher weights may work but I have not tested it, see XYZ_graphs). Increasing the weight of "arrow_in_body" may result in weirdly shaped arrows that may not even be stabbed into the person. See XYZ_graph for details. Increasing "arrow_\(projectile\)"'s weight also seems to increase arrow count but the effect doesn't appear as strong.

Lora mixing results vary. Most lora's work well with this one. ex. hanging, corpse. I did encounter a few lora mixing issues with some of my other Loras. It may be the case that Lora's which effect the body's shape/position may not work when mixed with this lora. That said this Lora DOES work with character lora's and does not destroy character prompts.

>>1604
I like it, but yeah, most of the arrows come out not looking like an arrow

weight 1.2 seems to be working best

other styles also work well

PissDrinking PonyXL lora
Author: Entropy (https://www.pixiv.net/en/users/32339993)

Triggers: drinking_pee
Supplementary: oral_invitation, open_mouth, tongue_out, own_hands_together, tongue, cupping_hands, 1boy, penis,
Weight: 1.0
Reccomended Epoch: 40

Download: https://mega.nz/folder/P2Rk1KAZ#siZJ8iBmWND6I4n555faFA


Notes: Creates an image of a girl getting pissed on and drinking it. Trained on images where urine is comming from a penis, results for drinking urine from a girl may vary (untested). Mixing with characters, styles and other concepts seems to work, but some character prompts/lora's clothing may overwrite the piss stream. Not using supplementary tags may result in an image where a girl is getting pissed on by a penis but no urine is collecting in the mouth (pissing may be not toward mouth, could be useful), it seems that "penis" generation is baked into drinking_pee.

>>1672
works with other lora's

>>1672
i really like this one. it is pretty flexible.
Notes:
1. Use strnegth 1, Itried higher strength, like 1.2 and imho I get worse results.
2. a very nice optional tag is side view.
3. I had trouble with the cupped hands lora so I never used it.

other artist loras work very well

a bitch should know her place

>>1674
It's not actually trained on "cupped hands". Most of what I put in the supplementary section is not a part of the lora. It's almost always a default pony keyword. Typically I put things in the supplementary section if A. it helps out with the generation of the main keyword (1boy, penis, etc.) or B. is close to or could be desirable (cupping_hands).

>>1689
fair enough. In my experience, despite years of advancment the AI still struggles with hands, especially with hands in strnage positions, and especially when an artist lora is present.

So I try to avoid hands as much as possible in my art.

I found a LoRA that imhho combines very well with the arrows one

https://civitai.com/models/490923/bound-wrists-up?modelVersionId=545896

I also tried with the punching bag, but results where very bad. this is th eonly one good enough out of 12

Pussy Stab PonyXL lora
Author: Entropy (https://www.pixiv.net/en/users/32339993), Anon (Image gathering), Anon (Thank you for the training data and sorting them into easily usable folders) Let me know if/how you want to be credited in the mega page.

Triggers: pussy_stab, sword, knife, rapier, katana
Supplementary: blood
Weight: 1.0
Reccomended Epoch: 12

Download: https://mega.nz/folder/Dm5iiDIA#KOjbKnX02qnPMdQ9hWUOQg

Notes: Creates an image of a girl getting stabbed in the pussy. Object used for stabbing depends on trigger. Trained on sword, knife, rapier, and katana as weapons. Lora weight at 1.0, modify trigger weights if prompt needs to be stronger.

Pussy_stab: Main trigger for lora. Should always be used. Use weight 1.0 or greater. Stronger weights positively effect stab generation quality until a certain point. I recommend using somewhere around 1.5. See XYZ graph for details, upper bound not yet found but upto 2.0 weight has been tested & confirmed working. Weight: 1.0+

sword: Weapon stabbing pussy becomes a sword. Higher weight creates better sword. Same deal as pussy_stab. Weight: 1.0~1.5+

knife: Weapon stabbing pussy becomes a knife. Weight: 1.2+

katana: Weapon stabbing pussy becomes a katana. Weight: 1.0~1.5+

Rapier: weapon stabbing pussy becomes a rapier. The weakest out of the 4 trained weapons. Generation quality is much lower than the rest. Consider avoiding unless you really want a rapier. Weights 1.5 and higher appear to cause some issues with the sword handle freakishly appearing where it shouldn't be. Starts growing on body parts like a tumor. Weird and kind of interesting to look at but not practical or useable in anyway (see xyz graph for examples). Weight: 1.0~1.5 (Avoid any higher)

Nice, btw is it possible that we get a proper beheading lora?

>>1800
damn! this is amazing. Thanks Entropy !

>>1803
Does the current one on the rentry not work? I'll look into making my own if there's demand for it.

>>1806
not that anon, but the view I tried from the rentry work so so

What i would like to see, is
!. a LoRA where a girl is just decapitated and you see her head flying near her body, possibly with a fountain of blood

2. a body with her head nearby, already dead,


I believe the decap that were trained used a mix of all decaps and have a bit of bleeding.

>>1806
For me, it's not working at all, in best scenarios there is some blood on the neck but that's it, no severed neck or head. I'm using Pony.

On a side note: we need to reorganize the rentry. Soem of the image are .png and the page take forever to load. We need to split it into 3 separate ones:
1.5
PonyXL
SDXL

Maybe soemthing with a drop down menu. Any alternatives to the rentry people cna suggest?

>>1800

Awesome! :D Thank you and thank you to Entropy, too!

Wish I could already use it, too, but I still haven't managed to get Pony running. xD

But it's really nice to see these first results here, though. Marvelous! :D Keep them comming if you like, please. Or if you'd rather not flood this thread please feel free to flood mine instead. :) (here: >>1538)

>Anon (Thank you for the training data and sorting them into easily usable folders) Let me know if/how you want to be credited in the mega page.

If it's not to much effort please credit me as "EeL" and by my pixiv, please. (I've not yet published stuff there but intend to in the future.)

Again, I can't thank you enough! Please let me know if there's anything I can do to return the favour!

>>1809
Whoops that might be my fault. What should I use instead so that I'm not significantly effecting load times?

>>1808
Alright I'll try making one. No promises on when it'll be done but I'll try.
>>1810
What's your pixiv? and Sure I'll flood your thread lol. Got any specific requests?

>>967 Almost a month later but I finally got around to it.

Eye penetration PonyXL lora
Author: Entropy (https://www.pixiv.net/en/users/32339993), Anon (Image collection)

Triggers: eye_penetration
Supplementary: blood, penis_on_face, cum_in_eye
Weight: 1.0~1.2
Reccomended Epoch: 35,40
Download: https://mega.nz/folder/auAFDAAD#lDDKhUsdc9Pich7uHXzxLA

Notes: Creates a penis penetrating an eye socket. Similarly with the eye socket lora, mentions of eye color or any prompts that effect the eye will make penetration less consistent. However, this lora works with high weights, increasing the trigger word "eye_penetration" will solve this issue without causing much distortion. Trigger weights upto 2.0 have been tested & seem to work fine (see xyz grid), I currently use 1.5~1.8 and it works well with mentions of eye color or character lora's that contain eye data. Increasing the Lora weight above 1.2ish causes the creation of less than ideal images. It's really just blurry and unsuable and some prompt data is destroyed. Keep Lora weight to below 1.2, I use 1.1 but 1.0 should work fine. I use epoch 35 but I honestly can say which of 35 or 40 is better.

Consider using supplementary tags as they significantly impact the end results. penis_on_face I would suggest as something you almost always want to use. Haven't tested it at anything other than a 1.0 weight but atleast at 1.0 it helps in generating a penis penetrating an eye.

Cum_in_eye causes cum in the eye. It's been slightly trained on so it's effects should be greater than it would be without this Lora.

blood causes blood to appear in the eye being penetrated, and on the penis penetrating the eye.

If you want a more clear image of just the girl using tags like disembodied_penis may help.

I also stopped uploading every safetensor file since I'm on the free version of mega and I only get 20gb. No point uploading useless files for the sake of having every single option. I'll also be deleting some of the safetensors in each Lora I've uploaded here that are "useless". So really low epoch safetensor files that I doubt anyone uses. They're like 100MB+ each so it'll save me some space. If for whatever reason you want them now's your one and only heads up.

>>1812
>What's your pixiv?
I sent you a pm on pixiv. :)

>and Sure I'll flood your thread lol. Got any specific requests?
Right now I'm just really curious to see what the lora can do. :D (But maybe some nice cuties in medieval/renaissance/fantasy dresses with corsets, perhaps? I guess you may know from the dataset that I really like those xD)

...If I could already use it myself I'd want to try and see what it can achieve with inpainting. (I mean txt2img with the lora and then img2img-inpaint to get get the swords/knives looking perfect) I can't wait! :D

>>1812

Oh, I just now realised that you've also made the pissdrinking lora. :o

If it's possible and not to much work I'd really love for some of those aforementioned corset wenchy maidens to get a nice drink/golden shower with their swordfuck, please! Can't let them die thirsty, now can we? :D

>>1826
not that anon. Here is my try.

>>1834

Nice! Thank you! :D

>>1807
Nice ideas, and I'll add one: it will be great if we can get an actaul good "severed head" lora, including different concepts such as head&feet in box, head on plate, head on spikes, head lifted by hand...etc. The ones we have right now is okay, but it's janky and got big problem producing things above.

>>1849
Don't forget detailed neck stumps.

>>1813
loving it

Since I've been playing with img2img a lot recently, I thought I could apply that to video2video. I tried a bunch of things but it ended up being that the less modifications I made the better. I could get fairly consistent img2img gens between frames but I found the result looking not too different then koikatsu gen I was using as a base for anywhere outside of the face. The result here is just Adetailer on an image sequence. Not too complex but I think it

I don't have means to generate my own, so requesting some more eye socket penetration pics. Bonus points for loli

>>2099
Here's one with Iori for now. Got any specific requests? Like character choice, clothing, setting etc. Or just anything if its eye pen.

>>2113
Any setting/characters works really, but here are some characters I like:
- Kokonoe Rin
- Rei Hino
- Konjiki no Yami
- Nagato Yuki
- Ana Coppola
- Hirasawa Yui
- Kinomoto Sakura
- Kuroki Tomoko
- Nagi (Kannagi)
- Hiiragi Kagami
- Shana
- Holo (Ookami to Koushinryou)
- Suigintou
Select if you happen to like any of these chars too.

Some where the girl is seemingly enjoying it would be nice as well.

Also thanks!

>>2236
ver with holding a holocoin if you know what those are. :)

very nice, many thanks!

you were creating amazing LoRAs

anything new in the pipeline?


>>2237
never heard of holocoins before, but I agree, it is a funny concept

>>2274
None at the moment. But I'm open to Lora ideas if you have any. Holocoins were a /biz/ thing on 4chan a couple years ago. Basically some anon minted a bunch of 1oz silver coins with holo on them.

https://en.numista.com/catalogue/exonumia319448.html

It's just a pretty cool trinket as a fan of spice and wolf so I thought I'd add it in.

>>2317
I got some suggestions.
now we have arrow wounds, how about gunshot wounds?
or some variations of hanging like shown here https://www.pixiv.net/users/28408963
or drag body like here
https://www.youtube.com/watch?v=XqJ2fmzoIuQ

Just updated the PonyXL section of the Guro Rentry
https://rentry.org/R-18G_Stable_Diffusion

It's been neglected for a couple months so I went and added in all the new Lora's made by the community plus fixed a few issues I made a while ago. Load times should be better now that I'm not forcing you to load tons of large images. Each image should only be about ~100kb now, at least for me load times are near instant now. My internet connection is acting up today and I keep losing connection while uploading Lora's to Catbox so alternate back up links for the new Lora's will have to wait for now.

Please let me know if I missed any community made Lora's or just good guro lora's in general. Also please let me know if you notice any errors. A good chunk of them aren't my Lora so while I tried to document them correctly it's always possible I made mistakes.

>>2319
I'll take a look at them I'm a bit preoccupied right now with another hobby project so I won't be pumping out Lora's like I did the last few months.

if you want ideas:

cut in half horizontally, two halves

giant circular saw cutting her in half

chainsaw cutting the pussy

harakiri

>>1578
BTW this guys work if you add ANytest

Been working on a guro game for the past week and a bit. It uses AI art so I figured I should post it here. Just wonder what the interest level in such a thing is if any. Making games takes a lot of effort and time so I'd like to know the general interest level before dedicating the next few months of my free time to it.

I can't upload videos so please check out my fanbox post (free): https://www.fanbox.cc/@entropy-game/posts/8194942

It has a short preview of my progress so far. Admittedly it's not a lot but should give you a rough idea of the "gameplay". It's basically going to be a bunch of guro scenes wrapped in a 3d world.

>>2594
And here's the one guro scene so far. I plan to make a lot more but this is all I have for now.

Any throat slitting lora?

>>2594
I like it a lot. the sprites have trouble rotating towards the viewer, which is annoying honestly. Otherwise is very nice.

I'm impressed you managed to have the consistency of the character.


Said this...

not many games born on this board had ausccess. And i think none of them gained money.

Smithyoversunset closed their site after a lack of interest.

other creators played for a couple of years and than disappeared.

I had an idea to make a game, but always thought it was not worth it.

making a game takes much more time than people initially plan.



So do it, only if you feel super passionate about it.


But I love the idea of another formal woman helping me torture the girl. Hope to see more.

>>2626
the solution to the sprite problem is the same as DOOM.

you have several version of the sprite rotated at different degrees.


But it takes much more work

>>2627
scartch this.

I rewatched your video seeing her gettin progressively more naked.

Spending tons of effrort to do the same sprite from different angles is not worth it.

just stick to this.

>>2626
If you're talking about how they don't always face the character it's intentional. They won't rotate towards you unless you are within a certain distance. As for the consistency you can thank Drawboto who told us about Anytest >>2068. Using Anytest with inpainting makes for surprisingly good consistency. I'm also using koikatsu models as a base which helps with consistency in the CG images.

I've made a game before so I know the ridiculous amount of effort it takes. Spent 2 years on one, released it on steam and got like 20 sales lol. I enjoy the process of creating so that definitely helps. I'm structuring the game in a way that will make sure it can be considered "finished" at any point outside of the initial work I'm doing right now. If for any reason I quit work on it there will at least be a complete package for people to enjoy.

>>2594

Like all your other stuff this looks awesome so far! Totally agree with >>2626. I have to admit though, I'm not that much into playing guro games, so my opinion shouldnt have that much weight. But if you ever need beta testers I'd love to give it a go! :)

I wonder if some form of 2Dto3D AI-conversion tool could be helpfull for your project. I haven't looked into these that much but I know they exist, so maybe that's worth taking a look at?

>>2668
It would probably be easier to use the 3D models I'm sourcing from koikatsu rather than to use AI to recreate them. I already export them to blender for rendering so it's really only one step away from getting them into Unity. Right now I'm doing the complete opposite by converting 3D models into 2D sprites with AI. pic related left is the 3D model render and the right is the AI generated result

There's this game called manabia (not guro)(https://www.dlsite.com/maniax/work/=/product_id/RJ321839.html) which is what I'm basing my game off of. The trailer for manabia doesn't show it too well but it has 2d sprites placed in a 3D world. The sprites will only rotate towards you once you are within a certain distance (as if to signify they notice you now). I wasn't a fan of it initially but this style really grew on me so I'm copying it for my game.

>>2669

Ah, interesting. Never saw a game with this concept before. Judgeing from what can be seen in the trailer and screenshots I guess the devs of that game worked around the 2d-problem (if one even wants to call it that, since its the style) by tunneling the player and extensive use of shadows. Do you plan on having comparably moody lighting and/or also narrowing the possible angles from which the player may encounter the npc's, too? :)

>>2671
I definitely want to play around with the lighting and view angles, but for now my focus is on creating a basic city area map and adding a few more characters. It's going to at least for now take place in the day, I may restrict player view by having characters in corridors/alleyways. I personally don't like the flash light effect. As for the moody atmosphere I don't think will match the game since I want to focus on consensual guro where both parties are happy. Thats sort of my primary fetish and is going to be the primary focus of the guro in the game. I think it will be going for a warmer lighting style for now.

That said I'm not going to even touch lighting until I can get the core of the game worked out.

I love the idea of a guro game mad ewith AI graphics.

The ability to choose the girls fate, to snuff her or to spare her. This control makes it super hot.

And if yo uare able to preserve the pipeline for each torture/snuff method you can realitvely easy remake the same scenarios with different girls.

damn I really hope you can bring this project to a functional prototype. I would gladly contribute and help in the development if I could.

Working on making Live2D animations for the game. Split up the AI generated image in photoshop and slapped it into Live2D. Not perfect but I think it looks ok. The idea to have this animation be user controllable. It will be very limited control but for this one my plan is to make it so that when you click you slam her against the wall.

>>2705
Most of the content will be specific to each character. I can remake the same scenarios but it won't take much less work, so I don't think I'll be doing it often. The plan is to have a playable version by the end of this month or early next month.

how many characters do you plan to include in the initial release?

>>2753
I wanted to include 4 but I think it's going to be around 2 or 3.

Quick little update on the game.

Overall the game is progressing well and my initial plan of early August release for the first playable version is still on track. However the game has been scaled back slightly and I plan to have 3 full characters on launch instead of 4. At the moment one character has been completed. Most of the development time so far has been spent on getting the basics implemented so I expect to be able to finish the other two quickly. Player selectable dialogue options for branching paths (will be small branches in the story for now but built to be easily expandable), interactable Live2D, sound effects, voice lines, and various backend controls to enhance the visual novel experience have been added.

This clip is the first execution scene I've completed in the game. I would appreciate it if any anons want to give some feedback on the audio, writing, and or AI images. Any feedback is appreciated, even if anything just feels wrong please let me know so I can work to improve it.

>>2865
i love it. Btw wesbites like eleven AI allow you to voice characters. You have free tokens each days.

I was able to get decent expostion just by writing text.

Stomping Live2D scene for executioners world game.

Progress is going well, and I expect to release an alpha of the game on Aug 3rd on Fanbox. Game will be fairly short for now, each character will take 5~10 mins and even less if you skip past the dialogue. It will have a simple introduction into the premise and let you out in to the open world. The introduction will have an execution so there will be only two other characters for the player to interact with. The two characters are going to be in pretty easy to spot places. But I plan to add some more hidden characters and really lean into the 3D open environment exploration later on.

https://www.fanbox.cc/@entropy-game/posts/8194942

A little comic I made for a request from pixiv. Nikusa from FnF, never heard of this character before so I have no idea if this remotely accurate to her character.

I'm loving the progress so far

Happy to say version 1 of the game is now available on fanbox. Ended up doing a bit of a crunch to get it finished for today so will be disappearing for a few days. It should take about 15~25 mins to complete everything (depending on how fast you read, and time you spend in live2D scenes).

https://entropy-game.fanbox.cc/posts/8343055

Please don't feel pressured into spending money if you don't want to btw. I know I've been shilling this for a while but I just want to spread my little project. For now it is paid only but I Will be posting a free demo some time this month (hopefully), or early next month, so please check it out then.

can train a guro lora with Mortal Kombat's fatality picture? this game has a lot of guro scenes

>>3067
I could try. Do you have a collection of images I could train on? At least 20 is required, high res preferably. I don't own any mortal combat games so collecting these would probably be somewhat difficult for me.

>>1813
Thank you for the LORA, anon, gotta start cooking later probably.

any chance you can make a circular saw LoRA?

Where a girl's pussy is near a circual saw .

Just peril, no need for guro.

>>3086
same deal as before, and this goes for everyone in the future as well. I will tag all the images and train the lora if you can gather images for training. At least 20 high quality images per concept/character/whatever, more is better but make sure they are high quality. I will simply throw out specific images if they are not high quality so just include the image if you aren't sure.

By high quality I mean.
- Decently high res (doesn't need to be super high res just nothing that's below 512x512)
- Shows what you want to be trained clearly.
- Style doesn't matter.

I would recommend putting them onto mega but whatever works, as long as I can download them.

I find gelbooru and pixiv the best for image gathering.

>>3099
I'll take you up on your offer. I can never get enough of taxidermy where the organs (especially the uterus) are on display outside the body. I tried training this myself but I couldn't get adequate results. https://files.catbox.moe/su7nej.zip

>>3322
Couple issues with the set. What are you trying to replicate here? You need to be a bit more specific imo otherwise you will end up failing which is likely what happened when you tried training on this.

You can do lets say taxidermy in glass or if you want a specific body part in glass (you would want 20 for each body part, you can have overlapping parts but each part should have 20 images that contain it)

I went ahead and collected the necessary images this time, but I won't be doing this every time. I'll be training this tmrw, it will be a taxidermy lora with a focus on uterus in glass, and taxidermy in glass / display. I will train on other parts too but they weren't a focus so I don't know how well anything other than uterus will turn out.

>>3322
Taxidermy PonyXL lora
Author: Entropy (https://www.pixiv.net/en/users/32339993), Anon (Image collection)

DOWNLAOD: https://mega.nz/folder/TrpxSaAQ#E4BoQaKboYq3oLX-XTAejw

Triggers: taxidermy
Supplementary: taxidermy in glass, uterus, uterus in glass, organs in glass, brain in glass
Weight: 1.0+ (1.3~1.5, maybe even higher)
Reccomended Epoch: 20

Notes: Creates an image of a taxidermied person. Due to lack of training data, supplementary tags do not work well. You may get lucky and have them appear but it seems to be very luck dependant. Changing the weights of the trigger words seem to do next to nothing? Honeslty I'm quite confused about this and I've never seen this before. Obviously something is wrong but I don't know. Increasing Lora weight will increase the "taxidermy"ness of the image. Characters will more often be attached to poles or be encased in a open box etc. I find using a Lora weight of 1.3~1.5 useful. Increasing Lora weight does cause the image to becomes more "washed out"... I'm not sure how else to discribe it but it kind of becomes more desaturated.

Might work if you try img2img but I haven't tested that out

>>3342
>>3322

I tried my best but I just can't get uterus in glass to work. I looked every where I could for more samples but only found 12 that worked. I'm sure there's more out there but I just couldn't find them. Maybe if img2img works I could retrain the model some time in the future. I like the idea and I appreciate the samples you gave me. Sucks that it didn't turn out too well but I hope this is of some use.

>>3345
I sincerely appreciate the effort. The lack of training data meant all my previous attempts at training it (on non-XL models) failed wretchedly. If I could find 40 images of uteri in jars I'd send it your way or learn to train XL models myself, but this is something niche within the niche of guro so I can't find anything worthwhile.

>>3099
I have another request: Self-harm scars or cutting scars. If possible, a multi-concept lora that can distinguish scar locations. I sorted the images into 3 folders based on where the scars were. Images with scars in multiple places appear multiple times.

If training multiple concepts into a single LoRA is too difficult, you can just merge them into a single folder. The 3 datasets share filenames so if you set them to overwrite each other and put them in the same folder you won't get duplicates. I tried to avoid putting too many images of Ame-chan in there to make sure it didn't become a character LoRA.

https://files.catbox.moe/wraqcl.zip

>>3359
Lora turned out pretty great. I'm doing some testing now. I'll post the Lora soon, probably later today once I've compiled all the information I find.

For future reference please don't use "sample" images if you're downloading from gel/danbooru. I know I said anything above 512x512 is fine but if there is a better version available I would prefer if you could get those instead.

I appreciate the separate folders but please make sure none of the images are overlapping, it just makes more work for me since I have to check the same image multiple times. Its fine if its just in one of the categories, I go through all the images one by one anyways.

I ended up merging upper arm and wrist cuts, so there's no way to distinguish creating those using txt2img.

>>3359
Self-harm scars PonyXL lora
Author: Entropy (https://www.pixiv.net/en/users/32339993), Anon (Image collection)

DOWNLOAD: https://mega.nz/folder/2mJlnIjQ#GnPxtlV-5w0Ylrq13LOmfg

Triggers: cuts, self-harm scar, scar on arm, scar on leg
Supplementary: injury, blood
Weight: 1.0
Reccomended Epoch: 15

Notes: Creates an image with self-harm scars on the arm or legs or both. Using the blood keyword makes the cut marks "fresh" and will have blood coming from them. Without it will appear as an older wound. Always use cuts and self-harm scar, you can try without it but I haven't tested that so can't say what happens.

scar on arm: for a cut on arm. Note that this doesn't make distinction between arm and wrist. You will get the cut marks wherever on the arm. Shouldn't be hard to clean up and get it exactly where you want (if you want it on the wrist or somewhere specific)

scar on leg: for a cut on the legs.

Epoch 20 is a bit fried but it kind of look nice? Has a bit of a phsyco style to it. Might be useful.

>>3370
Phenomenal work.

>>3370
is there enough training material to possibly add scars on the stomach in a future release?

>>3391
Future releases are not needed. The LoRA will add scars to the stomach if you use "Bruise on Stomach." It seems to have made an association between many forms of injury and cutting scars and only knows cutting scars do not belong on the face. Specify an injury to a body part and you can get cutting scars there.

>>3393
As an example, the dataset had 0 examples of cutting scars on breasts but the LoRA can produce scars there.

>>3393
>>3394
Wow I didn't even know it could do that. Gotta hand it to training magic. Doing shit I didn't even add.

Dear Entrophy.

can I request a character LoRA?

Princess from Angel COrps

https://files.catbox.moe/tqhv0w.zip

I compiled all the art I found of her.

Please she needs to coem back to this site.

>>3564

not Entropy obviously, but just in case it helps, there's already a Lora for 1.5 for her: https://civitai.com/models/528348/sammie ...If nothing else maybe the preview pics are good enough to add to the dataset.

But also there are 2 TehDave-Style Loras. Just by using the triggerwords of the sammy lora with https://civitai.com/models/318858?modelVersionId=642597 I get stuff like these pics on the left. I'd say that's pretty close! :D

>>3564

There's also https://civitai.com/models/497156?modelVersionId=552664. Also very close I'd say.

And in case you like the style, these were both made with a mix of 0.5 https://civitai.com/models/500876/icatcher-cartoon with 0.5 https://civitai.com/models/500858?modelVersionId=556730 and then the result of that mixxed with 0.3 https://civitai.com/models/632040/cutecore-xlb59

(check meta for prompt)

>>3564
Just saw your request. I'll work on these fairly soon, just a bit busy rn so can't say exactly when they'll be done.

>>3571
>>3573
>>3570

THank you. The body and clothing transmits very good. But the face and hair are still off.

>>3577
decreasing the weight of both models below 0.4 and removing chubby helped

>>3577

>>3578

These are both lovely! But I must admit, that I barely see the difference on the face. The hair stands out more but still, if you'd have shown me these 3 years ago and told me these where works by dave himself I would have believed it. But the same applies to this pic here... Even though the styles look very different. Guess I really don't have an eye for that sort of thing.

>decreasing the weight of both models below 0.4 and removing chubby helped

Thanks for that hint, I hadn't even tried mixing both Loras at lower weights by then and thanks to you I didn't have to experiment with the weights for too long. 0.3 works just fine for me. Just like with the styles I've used both your hanging lora and the hangboard lora at lower weights for these 3 pics here.

...this was done prior to that with just the TehdaveV2 Lora at full weight...

...aaand lastly this one was done with all three style Loras at 0.3.
Yes, all three, because someone just dropped a third TehDave style Lora: https://civitai.com/models/738490/psys-pony-style-loras

this one is clearly overtrained but it was definitely trained on angels corps girls. Although princess is mentioned only 3 times and the tag syntax was wrong. (check Lora meta and you'll see what I mean. And don't use prompt "angel corps", you'll just get angelwings. xD)

Hope you don't mind me posting these 3 pics here. I don't want to completely spam and derail your thread, but I thought this was relevant enough and therefor warranted. (if nothing else at least these 3 pics also showcase your hanging lora. Hope that counts^^)

I know the faces and hair are off in all three but I really have no good eyes for that. So I'm also curious which one (if any) is closest in your opinion. :)

>>3590
>>3588
Damn this is hot.

Can I ask you to do one pic of her without the iCatcher LoRA?

>>3591

iCatcher Lora? That's not a Lora. That mix is my base model checkpoint. :D

But I can try another checkpoint if the whole style is what you mean. >>3577 and >>3578 were done with baseline PonyXL_V6 if I read the meta correctly. You want my to try that style?

>>3595
yes please.

>>3597

Well, I tried, but it turns out I can't really handle basic pony_v6. It behaves so differently from what I've used until now, especially weight wise. I can't wrap my head around it! None of my prompts work as intended. "Backgrounds? What's a background?" The initial gens all look like prehistoric cave paintings, scribbled by a dieing neanderthal with late stage parkinson... Even simple shit without a lora... Tried the whole night but this is the best I could come up with. The likeness and style are still miles off, I know.

>>3603
that..... Is actually a very important thing you said to me.

I never tried other tings beside the default Pony.

Myabe I should Try another one. Thanks!

>>3564

Princess (angel_corps) lora
Author: Entropy (https://www.pixiv.net/en/users/32339993), Anon (Image collection)

DOWNLOAD: https://mega.nz/folder/OygD1ToC#sW_4GTfvcnaxRgWEDwRTKw

Triggers: princess_\(angel_corps\)
Supplementary: freckles, long hair, brown hair, green eyes, glasses, gloves, tiara, bracelet, fingerless gloves, armlet, ponytail | slingshot swimsuit,

Weight: 1.0
Reccomended Epoch: 15

Notes: Recreates the character Princess form angle corps. Works with other loras. Clothing is not baked into the main trigger for the most part, however certain clothing tags have traning on them like slingshot swimsuit.

pic related is made using a hanging lora

>>3617

Sorry for the delay. Was busy with my other project. It does work with other clothing triggers. Haven't tested it extensively so can't say how well it would all work, some triggers may have data I don't know about.

>>3564
And just to let you know, for next time please don't include full comic pages, and images with resolution below 256. Full pages have too much non-relevant information that could make it hard to train on, and images with resolution below 256 just don't work.

>>3617
that is awesome!

THank you so much!

>>3622
>>3621
It's awesome

Also I think the weight can be brought down to 0.7 without losing to much of quality

>>2408
I notices some guro loras on ehentai (https://e-hentai.org/g/2928445/7da13650d4/)
maybe you can upload them on rentry?

If it's alright to use this thread as a sort of sounding board/venting space for guro concepts that I can't yet fulfill -- in recognition of your excellent work in making so many possible -- the one thing that I would kill for (metaphorically!) would be a lora focused on broken limbs and/or joints bent in the wrong direction. Maybe other people have some tips for managing this with existing assets?

>>3816
I'm with you on this. although I'm prob more of a sick fuck tbh. I wanna see exposed bone, like from a forearm snapped in half. Also, would like to see ANY male focus.. Bout to lose my lunch over all these oversized tits..

Requesting a Pokonyan or Burn at the stake LoRa

Mate, your LoRAs were amazing.

Any chacne you have been working on anything cool and just have not shared it yet?

Dear Entropy,

I have a sorted dataset for Flux, any chance you are willing to caption and train a model fro FLux?

I took the liberty of rearraning the repository.

https://rentry.org/R-18G_Stable_Diffusion

I created section for each individual Model family subster.

You can access and modify each subset as before with the same password.





Now it has a section on FLUX.

>>4008
Thanks, I've been super busy irl lately and haven't had much time to do anything Lora wise but I have been working on a game.

https://www.fanbox.cc/@entropy-game

>>4007
As with all requests I will do them if you can provide me an image set of at least 20 images . I'll probably create a new thread with my requirements for requests clearly laid out so it's always at the top.

>>4009
I could try? I have never worked with Flux before so can't say if I'll be able to get any results. You can dump a link to them here or to my pixiv.

>>4061
Thank you for this. I've been neglecting the repository for a while now. Like mentioned before I've been busy af this month and didn't have time to fix that up. Looks much nicer and more organized.

no one's made a branding iron/hot metal/burning torture lora yet? i've been trying to get that to work with keywords in illustrious but no luck so far.

>>698
Solid after a little inpainting