//h/
File: 1759282223335139.png (1006.1 KB)
1006.1 KB
Previous Thread: >>8881901

>LOCAL UI
Forge Neo (ANIMA SUPPORT): https://github.com/Haoming02/sd-webui-forge-classic/tree/neo
reForge: https://github.com/Panchovix/stable-diffusion-webui-reForge | Stable commit: 6964ceb
Comfy: https://github.com/Comfy-Org/ComfyUI | https://comfyanonymous.github.io/ComfyUI_examples
SD.Next: https://github.com/vladmandic/sdnext

>ANIMA
https://huggingface.co/circlestone-labs/Anima
https://thetacursed.github.io/Anima-Style-Explorer

>NOOBAI-XL
https://civitai.com/models/833294/noobai-xl-nai-xl
https://huggingface.co/Laxhar

>NOOBAI SHITMIXES
102d custom: https://civitai.com/models/1201815?modelVersionId=1491533
291h: https://civitai.com/models/1301670/291h

>PAID
https://novelai.net/image
NOTE: Do not feed the trolls. Don't take discussions about NAI too seriously.

>RESOURCES
4chanX Catbox/NAI prompt userscript: https://rentry.org/hdgcb
Easy booru tag export userscript: https://github.com/Takenoko3333/Danbooru-Tags-Sort-Exporter
Tags: https://danbooru.donmai.us/wiki_pages/tag_groups | https://danbooru.donmai.us/related_tag | https://tagexplorer.github.io/#/
Inpaint: https://files.catbox.moe/zy2xlr.jpg | https://huggingface.co/Wenaka/NoobAI_XL_Inpainting_ControlNet_Full
Upscalers: https://openmodeldb.info
Booru: https://aibooru.online
ControlNet: https://rentry.org/dummycontrolnet (OLD) | https://civitai.com/models/136070 (look at the green links in the description)

>TRAINING
Guide: https://rentry.org/yahbgtr (WIP)
Anon's scripts: https://mega.nz/folder/VxYFhAYb#FQZn8iz_SxWV3x1BBaJGbw
sd-scripts trainers: https://github.com/67372a/LoRA_Easy_Training_Scripts (check out refresh branch) | https://github.com/derrian-distro/LoRA_Easy_Training_Scripts | https://github.com/bmaltais/kohya_ss
Other trainers: https://github.com/Nerogar/OneTrainer | https://github.com/tdrussell/diffusion-pipe

OP Template/Logo: https://rentry.org/hggop/edit | https://files.catbox.moe/om5a99.png
Showing all 2050 replies.
>>
File: 1753889280122815.png (2.7 MB)
2.7 MB
>>
>>8887702
How can she be so bored and so sweaty at the same time?
>>
>>8887705
Because she's a lesbian and she's struggling with imagining it's a girl with a strapon
>>
File: requests.png (3.4 MB)
3.4 MB
Request anchor :D :D :D :D :D
>>
>>8887705
She's currently in Europe
>>
>>8887708
this is some early ayakon tier shit
you really don't have to prompt for high heels and gloves when she wears them by default, they'll just end up on everyone else
>>
>>8887705
it's like that meme where the guy is trying not to fart next to his crush. she forgot to poopoo before anal sex
>>
File: 1723889280122835.png (916.8 KB)
916.8 KB
>>8887699
removed the onions
>>
File: prodiggyconstant.png (43.8 KB)
43.8 KB
>>8887503
I'm gonna need peer review on this before I bake, it'll be 110 epochs, 40 images as I don't know how long it'll go.

Thanks for the daily help guys.
>>
File: 1756087186024207.png (2.5 MB)
2.5 MB
>>8887702
>forgot to remove the turbo lora from the prompt after deeming it shit and reverting the rest of the settings
...
>>
>>8887718
i assume this is the one with the turbo lora?
>>
Realistic tags are actually kinda helpful with anatomy, background and composition, particularly if you're not going for a pure animu style
Maybe tag prompting will be useful
>>
>>8887719
No, the VERY painterly one from earlier is. This is the intended look.
>>
>>8887722
wow
>>
Why is having low shift a bad thing
I've seen anons talk about it in these threads but in my personal tests it looks better than having higher values
Has anyone done a x/y/z plot on this?
>>
>>8887716
For ProdigyPlusScheduleFree, delete all optimizer args except weight_decay https://github.com/LoganBooker/prodigy-plus-schedule-free/blob/main/prodigyplus/prodigy_plus_schedulefree.py
It's a schedule free optimizer so just set the epoch to something like 1000 and test every epoch until you get a good epoch then stop training
>>
>>8887728
thanks, and if I were to do normal prodigy, it'll be like that setting I shown?

1k epoch is i'll assume 24 hours though.
>>
67
>>
>>8887729
You don't need to train for 1000 epochs. Train until you get a good epoch
>>
File: 1748452519245052.png (2.7 MB)
2.7 MB
>>
The new turbo lora is ass
>>
>turbo lora is ass
you do not say
>>
>>8887734
i'll bake straight away up till 90 epochs like usual and see the findings, it's not like I can stop it and then gen something so I must see it to the end.
>>
>>8887736
I agree
The only upside is that it makes inpainting easier somehow
I suppose it's a good thing to have in handy when you want to test composition really fast
>>
>>8887734
nta but is there a convenient way to do that? save state, try out the lora then resume, and manually repeat as many times as it takes?
>>
File: sav.png (19.9 KB)
19.9 KB
>>8887741
I have it ticked in now hope it works.
>>
>>8887741
Yes, saving state and then resuming is the way
>>
>>8887742
iirc it takes a ton of disk space, I wouldn't save after every single epoch just pick a place to stop and save there
>>
>euler_a: Softer, thinner lines. Can sometimes tend towards a 2.5D look. CFG can be pushed a bit higher than other samplers without burning the image.
Yet retards in here will pretend euler a is bad
>>
>Can sometimes tend towards a 2.5D look.
blud. your own quote just explained why euler is SHIT
>>
Euler A is not bad, it's ASS
>>
>>8887746
a
>>
>>8887746
What do you mean 2.5D is peak
>>
>>8887746
I wonder what they meant with that part. It smooths out gradients, but you don't have gradients in the first place in actual 2D. Could it be referring to 3D stuff?
>>
>>8887745
euler A is shit because it doesn't work with controlnets if you use them the way you're supposed to use them and it's shit for upscaling.
genning straight t2i at higher res it's "fine." not really great or preferable but it's "fine."
>>
>>8887751
sounds like bullshit, upscaling works fine too
and how are you not supposed to use controlnets
>>
>needing to use controlnets
literally just write nigga it works every time
>>
>>8887754
>just write
I'm brown
>>
>>8887751
imo depends on the style you're going for
I know people here hate anything that isn't anime coloring, flat color, (oekaki:2.3) but euler a 6 cfg works nicely with soft shading and some artist tags
>>
Any recommendations for what LTX model to use with a 5080 and 32gb RAM?
Mostly just looking to do i2v
>>
>>8887757
use the english dictionary csv for tag autocomplete
and search for stock images they usually have a description you can copy
>>
>>8887760
LTX is abysmal dogshit for anime, it has absolutely zero idea what it's supposed to look like and will make it into uncanny 2.99D aberrations within just a few frames. Use Dasiwa (WAN 2.2 shitmix) with the general NSFW LoRA if needed. Layering LoRA on top of LoRA really fucks with the details so you should only use the bare minimum, which with an NSFW-capable shitmix is just that one LoRA for when you just aren't getting enough motion.
>>
File: 1764827456508109.jpg (411.0 KB)
411.0 KB
>>8887759
>>
>>8887760
the daisiwa one I tried just like this guy: >>8887763
It takes me about ~2-3 mins just to gen but I have a 4070 ti + 64gb ram though.

Also that guy's base workflow is bloated with unnecessity things tf you need voices for, etc. but at least it works.
>>
What's up with the shift next to cfg when using anima. It's set to 3 by default, should I change that?
>>
>>8887764
cool lizerd. box?
>>
>>8887766
nobody knows what it does, like clip_skip back on SD
some low-level shit that's supposed to make the model behave "different"
>>
>>8887769
clip skip isn't even relevant anymore but it's still treated like it is by some people.
>>
>>8887753
>upscaling works fine too
if you understood how euler a worked you'd understand that it works in a technical sense but it's pretty bad for it.
>and how are you not supposed to use controlnets
euler a only really "works" for controlnets if you have them on for the full generation duration. which you shouldn't do. Ideally you have controlnet end at 40-60% step count but if you do that with euler A it'll basically just erase what you got from the controlnet and do its own dumb bullshit rather than properly building off of it.
>>
>>8887772
It does nothing on architectures other than SD1.5 (and possibly old SD1?). But some UIs fill it into metadata, Civitai asks for it when uploading pics, etc.
>>
>>8887772
iirc it's just cumragsoft that cares. sdxl needed it flagged to -2 or something to work properly? I think. but on any a1111 based UI, clip skip does absolutely nothing unless you're using sd1.5 based models.
>>
>>8887778
Not SDXL, SD 1.5, more specifically the NAIv1 leak.
>>
>>8887766
I don't know the proper explanation but in my experience it's basically where in the gen it distributes most of the noise
If you set it to the lowest, it will immediately gen most of the details and the artstyle very early on, and then it will slow down as it refines it further
If you set it to the max, there will be a lot of meandering noise early on and it will slowly build up the details while accelerating towards the last steps
You would think this is straight-forward, but not really. Setting it at low will make the artstyle and the details look better but can fuck up your anatomy and composition and vice-versa, plus this affects how you use prompt editing greatly
>>
>>8887767
Which one
>>
>>8887783
could you send some examples, maybe an xyz plot? I can make my own in a second but I'm running another experiment
>>
>>8887786
i like both
>>
>>8887783
Due to how distributing works, if you set shift to high things like
[@artist A::0.8], [@artist B:0.8] will still have a relevant effect in your gen, but setting it to low will only make 0~0.3 or so actually relevant
>>
>>8887778
-2, zero, or nothing
Anything but 1 really, because that's when it tries to enable clip skip and it's not meant to be used with models other than SD1.5. Webui/Forge prevents you from being retarded and just ignores the setting no matter what you put it.
>>
>>8887787
Sure, but it will take a while my hardware is kinda bad
Will be a good experiment since I'm using an autistic artist mix with prompt editing
>>8887788
https://files.catbox.moe/zyzp1u.png https://files.catbox.moe/65wj4g.png
>>
>>8887782
does cumragsoft use - as a denotation or something and not as "negative?" because yeah nai leak/subsequent models used clip skip 2, but I mean that cumsoft breaks SDXL if it's set to anything other than -2 it would break, as a sort of function of the software itself.
but no, sd1.5 models used clip skip as a sort of "layer" function. There were some really funky fucked up furshit models that tried to contain specific features/datasets to individual clip skip values.
>>8887791
yeah that makes sense. And yeah, a1111 ui's completely disable it for anything not sd1.5, since it doesn't work/isn't intended there to begin with so there's no point in allowing anyone to enable or use it.
>>
>>8887792
You and me both. My poor 3060 has been through so much.
>>
>>8887774
That makes no sense. You've already pushed the gen elsewhere than the seed would have converged to, so when you cut off guidance both samplers continue operating as if you'd just started an img2img with partial denoise from that point.
>>
>>8887795
I have a 3050 with 8gb vram lol, the average gen with 30 steps takes a min and 20 seconds or so
It is most unfortunate that a x/y/z plot doesn't accurately represent what goes on with shift since you can't visually see noise throughout the gen like I can since I have live preview on
https://files.catbox.moe/d73tb2.jpg
Check the artist mix to see the prompt editing
>>
send nudes
>>
File: 1764822513445633.png (2.5 MB)
2.5 MB
>>
>copy good anima gen prompt
>change one setting or tag slightly
>suddenly shitty gen
Is it just me or this shit finicky
>>
>>8887805
remember noob base model?
>>
Maybe I should start posting some of my slop gens here after all. Thread has so few image posts these days it might still be fun to see for some people around here.
>>
>>8887702
slop

>>8887718
SOUL
>>
>>8887815
do it faget
>>
holy fucking shit people in this thread do not know what is clip skip anymore
>>
File: ComfyUI_41814_.png (1.1 MB)
1.1 MB
>>8887822
Is it that surprising? Any oldfag who's been here since pre-SDXL days you would be able to recognize from their posting or gen style, it's been years.

>>8887815
This is the only way forward. Though you can't revive the thread on your own, you'd have to get other people to genpost as well. Also should probably do it in /hdg/ and end this farcical split, it's gone on long enough.
>>
>>8887797
>both samplers continue operating as if you'd just started an img2img with partial denoise from that point
and what do you think euler a does?
oh right it tosses more noise on top of it and you just lost everything you forced it to try and do(since that added noise allows it to try and correct itself back to what it WANTED to do), while regular euler will be forced to build off what it already got to.
>>
>>8887830
It doesn't add additional noise on top of what was already denoised with the controlnet's help. Those steps are done.
>>
>>8887831
>It doesn't add additional noise on top of what was already denoised with the controlnet's help
it does, in fact, actually do that.
>>
>>8887833
This is a strangely common misunderstanding of what ancestral samplers do. At each step they remove some of the remaining noise and replace it with a new random distribution. The denoised potions of the image remain untouched. The practical result of this is they converge to a different result whenever you change the step count, because that changes the noise schedule it pulls this "random" distribution from.

You can check this yourself as well, using Comfy's Advanced Ksampler. Stop the gen with a couple steps remaining and decode the result, you will see the already denoised portions do not change as it advances through further steps.
>>
Thoughts on the anima turbo and highres aesthetic boost loras? I usually stay away but this time around they don't seem all too bad.
>>
>>8887716
SchedulefreeProdigy isin't as good as came/rex combo, this one repeats certain hair waves, higher epochs shrink the head while elongating/widening the body, including adding the hollow circular highlight I don't even have on the sameples unless I increase the lora weight as high as 1.5, came/rex will just either do one of the two things amost all the bakes, I'm reaching the goldilocks zone for came/rex anyways, but doesn't necessarily mean I won't test other optimizer/training combos.
>>
File: shift.png (31.0 KB)
31.0 KB
>>8887766
shift influences how quickly the model removes the noise per each sampling (or training) step, with lower shift (e.g. 1.5) you almost immediately skip to low-noise steps, meaning composition is going to be ass because you gave very little steps to it. with higher shift (e.g. 7.0) you spend more time in high-noise region.
also note that shift is the only "correct" way to adjust sampling schedule of flow matching models, "beta", "karras", "exponential", "ays" or whatever were not made for flow matching and when you use them there's a ton of math fuckery going on without a good reason.
>>8887769
clip skip is a hack introduced in novelai's first anime finetune of stable diffusion (they resumed from sd 1.4), basically it cuts some last layers of the text encoder before sending embeddings to the diffusion model. supposedly the model has richer embeddings to work on that way. sdxl adopted this trick for both of its dual clip text encoders from the get go iirc, and you don't need to set it manually.
>>8887745
>>8887830
the only thing euler a does differently from normal euler is that it removes the noise it initially added and then back puts the same amount of different noise. sometimes this may fuck up sampling with weird sampling tricks. maybe it may fuck something up if the model was trained with, for example, optimal transport (which tries to avoid training on very very bad noise in order to make the model rely on noise more) or some other noise energy bullshit. i'm highly surprised but this anon is 100% correct >>8887840
>>
>>8887847
>back puts
puts back
>>
The nudes?
>>
>>>/e/
>>
You fucker motherfucker
Once i get back to my home you'll see
>>
bloody bastard sir I beat your families cow
>>
gift card idea: chudai aunties
>>
little brown bro thinks he's all entrepreneurial
>>
the great thing about being indian is I only need to sell 2 AI images a month to make the average wage!
>>
No lies spotted
>>
Unlike SDXL, Anima uses the Qwen text coder, offering the significant advantage of being able to output images with the intended composition using natural language prompts (such as images with multiple people or composite scenes).

If WAI's image quality and Anima's prompt understanding are combined, it will become the standard model for the future.
>>
little brown bro tried
>>
>https://files.catbox.moe/cg2ucb.png
>>
>https://x.com/hulixiaobai/status/2059306763741266196?s=20
>>
>https://x.com/rushicrypto/status/2059360925787046174?s=20
>>
File: file.png (1.1 MB)
1.1 MB
>>
>>8887893
yeah AI let's me use my coding skills to make art, a hobby I always liked but never had the time to get "good" at because I have to afford rent.

>your best was good enough
no it wasn't I grinded shitty drawing for years I was never going to be good. Although it does leave me annoyed at people I am far better at doing traditional art complaining about AI stealing their work because they don't understand the tech,
>>
lul
>>
>https://x.com/firedotalt
>>
bros.. how do I deal with knowing this kimchi slut will never remotely admonish my little brown existence with such unbridled contempt while draining me into a prune with her grippers?
>>
wut cfg best for anima
>>
not too low
not too high
>>
File: 00033-4191145263.jpg (723.7 KB)
723.7 KB
>>
>https://civitai.red/models/2621858/realanima-x?modelVersionId=2980523
I found the Preview 3 version to be excellent for cg vidya sloots.
>>
>>8887934
why not base1
>>
>>8887915
bout tree fiddy, ironically
>>
>>8887937
Haven't tried it yet but it's what I linked. Just sharing P3 was p good so hoping base1 is better.
>>
(chibi:-1.8) in positives with negpip
>>
File: ComfyUI_29598_.png (3.5 MB)
3.5 MB
>>
>>
2.9 MB
>>
File: original[1].png (1.1 MB)
1.1 MB
here's your local novelai reference, bro
https://www.reddit.com/r/StableDiffusion/comments/1totumo/anima_can_edit_images_and_this_is_possible_in_two/
https://civitai.red/models/2654416/anima-can-edit-images-or-testing-anima-edit-loras-and-ic-methods?modelVersionId=2980586
>>
>>8887950
P-pancho.. *cough* forge neo support onegai.. *cough*
>>
>>8887952
forge neo isn't panchos, you mongrel
>>
>>8887953
Yes, nigger. I'm aware. But HochiMing02 isn't thread culture.
>>
File: t202.jpg (3.4 MB)
3.4 MB
>>8887883
>>8887930
Give SDXL some rest bro...
>>
cucks
>>
ducks
>>
bucks
>>
>>8887958
anima sucks doe
>>
yep yep yep many are saying this
>>
anima desperately needs e621
>>
File: 1769920818623633.png (774.1 KB)
774.1 KB
>>8887950
So he just described the NAI inpaint trick from the pre-reference times and then highlighted two nigh unusable LoRAs extremely limited in what they can do.
>>
>>8887971
I guess the main takeaway is just that the inpaint trick works with the Lllite inpaint controlnet.
>>
>>8887972
I thought anons here established that shortly after the LLLite Inpaint model came out?
>>
>>8887972
old news >>8882511
>>
File: Cynthia_Rape.png (1020.7 KB)
1020.7 KB
>>
File: out.webm (2.7 MB)
2.7 MB
>>8887971
>trick from the pre-reference times
The trick is still largely relevant and wasn't replaced by char reference. For example reference works like shit with inpainting model (they even have a warning for it) + anlas cost is gay.
>>
File: Ts_00030_.png (548.3 KB)
548.3 KB
>>
File: catbox_4p8z48.png (3.3 MB)
3.3 MB
>>8887894
catbox? I really like the style
>>
File: erza_00043_.png (1.7 MB)
1.7 MB
>>
>>8888003
cry
>>
File: 00074-3515761564.png (3.0 MB)
3.0 MB
>>8887958
use case for not using SDXL?
>>
>>8888020
anything besides 1girl standing. but we don't do that here
>>
>>8888020
I want a robot gf so bad, bros...
>>
File: 00026-1818123197.png (2.1 MB)
2.1 MB
>>8888024
>anything besides 1girl standing
any example?
>>
>>8888026
nope sorry can't think of anything
>>
File: 00003-2551090214.png (3.1 MB)
3.1 MB
>>8888027
based
>>
>>8887958
ani said that anima is heavily shilled by comfyanon's matrix server. the hype is almost entirely artificial
>>
File: t203c.png (3.0 MB)
3.0 MB
>>8888020
New = better.
>>
>>8888037
can you get a new brain then pls
>>
/asdg/ seems gone for good >>>/aco/9208012
Likely a follow-up to /r/ deletion which was due to deepfakes.
>>
>>8888046
good riddance
>>
>>8888048
Losing a containment thread hurts everyone else.
>>
>>8888046
can't they just move to /s/ or /gif/ or /b/ even
>>
>>8888046
I was going to say they missed one >>>/aco/9196124 but in the time I was making this post it got nuked. oops.
(also I legitimately do not actually know what the difference between that thread and asdg was at all)
>>8888050
only if the content it was containing isn't banned sitewide.
if it's banned sitewide, losing a containment thread doesn't matter - there's nothing to contain.
>>
>>8888053
/asdg/ was local, mostly Chroma, Z-Image and SDXL merges. MMAI was about trying to trick various SAAS options into doing nudity.
>>
yeah but good anima gens doko
>>
>>8888055
sounds like a dumb and worthless endeavor.
also it seems that their futa thread is mostly realism so that's kind of... funny. wonder if that'll also get nuked.
>>
>>8888057
>>8887958
>>8888037
>>
:skull: :skull: :skull:
>>
when will /h/ finally ban 2.5d bullshit? if /aco/ can remove realistic crap, so should /h/
>>
p*interlybros, our response?
>>
>>8888058
No big dead, it's been slowly dying for like a year anyway. I would put up a gachaslut cosplay once in a while.
Just feels like the site is getting shittier every other day, and this is another step.
>>
if only there was another site without these issues which already has bunch of cool people from the thread postin :^)
>>
>>8888066
isn't treechan dead? the link didn't work last time I tried
>>
every alt-chan is always dead, come join our thread with 1 post a day
>>
>>8888063
Why would they ban the only hentai that looks good? Nobody wants to see primitive doodles made by unskilled losers, much less masturbate to them.
>>
>>8888070
shut the fuck up and consider suicide you faggot
>>
Bery angery...
>>
i love masturbating to doodles
>>
>>8888046
Never went there, but I assume they started to generate unironic pedoshit.
>>
>>8888086
wrong, 4chan is just getting rid of "deepfakes" and 3dpd ai in general
>>
nice. very dead.
>>
File: 00224-2953476747.png (2.2 MB)
2.2 MB
https://files.catbox.moe/3oa34h.png
https://files.catbox.moe/otqa35.png
>>
weow
>>
i miss fernus
>>
File: 00242-3524458080.png (2.4 MB)
2.4 MB
>>8888107
>>
sir, where are the big heads, can't shitpost about that shit
>>
>>8887819
>>8887827
Alright, you've convinced me. I'll post some of my slop gens.
>>
File: 1769327660022143.jpg (331.5 KB)
331.5 KB
>>8888107
>>
>>8888113
welcome back, i've been extensively doing different loras lately but considering your more used to gloras, (to an extent 100% of my noob loras was based on your grand design since noob 0.5.), how do you combat with overfitting? the color, body is near on spot as it can be but the moment I do "from side, all fours, etc" it'll do exactly how the body is (i.e one hand lifted up, one hand under breast) despite not adding in the prompt, the one i'm doing is the came/rawr combo lately in the past few days never had that problem.
>>
>>
>>8888113
I know everyone will have different scaling of images with 4chanx so I'll just screenshot mine.
Do you see the grid pattern when on a zoomed out version of the image?
This exclusively happens with anima. Basically when it generates any kind of grainy/noisy small texture it repeats in a patterned way and a crapton of images look like this zoomed out. Basically why I have to try my best to generate the cleanest shit possible with anima and add noise myself if I really want it.
>>
>>8888125
on my screen only the fern one does that, your screenshot right now and the original, i'll assume the type of texture/grid stitch a real life canvas has.
>>
File: 1772195656822319.jpg (431.3 KB)
431.3 KB
>>
>>8888127
Real life canvas will not have imperfections that perfectly repeat themselves every square cm. And it happens with non-painterly images too.
>>
>>8888125
the grid thing is absolutely something I'm aware of/have known about, and I notice it when checking the images with windows photo viewer when it's not zoomed in on a 1080p monitor.
it seems like it's just an anima thing? It happens on "full background" outputs.
this also isn't dataset dependent. I've seen it happen on other datasets I've trained on. It also doesn't always seem to exist, it just happens sometimes when you output at larger resolutions. It's also possible it might be associated with one of the media tags? would need to test for it I guess.
>>
>>8888131
Yeah it's not a dataset problem because for me it often happens without any loras. Can also happen with just a single artist with no quality tags and media tags at all. Maybe bigger resolution triggers this more but not sure.
It seems that for grain it just learned one pattern that gives it "optimal" score on average and then it just keeps using it for every single latent pixel?
>>
File: animashit.png (86.3 KB)
86.3 KB
>>8888123
This is the setting I use with glora, again very good body, color correct. However, ~90% of the seed chance the position will have the same posing, camera angle, and even sometimes missing a limb. came/rawr doesn't do that, i've tried bat 1, bat 2 and now bat 4 already.

Either way it's kinda fun baking, i've made an extensive way to test loras and grade checkmark it.
>>
>>8888138
The bat 4 one starts to have less of the "same position" thing i'm having problems with but regardless it still does it.
>>
>>8888138
does kohya have a bug on amd where it does ~20s of idling after each epoch?
>>
>>8888142
I have an amd cpu but nvidia gpu, also i'm using the ezscripts since kohya's ass gui never updated to anima.
>>
>>8888131
oh correciton, even with a white background it does show up, it's just much more faint <https://files.catbox.moe/efykdu.png>
>>8888137
yeah I think it's caused entirely by anima(or possibly just being nvidia cosmos as a base) and it seems to be related to anything that's kind of textured but doesn't seem to exist much on flatter colors. Though, funny enough, this is the same lora, same seed, same prompt but with a different artist call(kiato is in the dataset for this lora) but doesn't seem to produce the gridding <https://files.catbox.moe/widlv5.png>
raw output <https://files.catbox.moe/djju1p.png>

>>8888123
>>8888138
oh right sorry I kind of had to do a lot of random shit.
Anyway; anima seems kind of wonky/different. It does actually get a lot more awkwardly overfit than I'd like and have been trying to do a bunch of different things to try and combat it.
Upping batch size does work to a degree but it also seems to hit reproduction a lot harder and the LR's. I've been doing a lot of random experiments/testing and so far anima does actually do a really solid job of tag differentiation for styles, which you can kind of abuse to do a sort of soft form of regularization by including a dataset from a different artist at a lower repeat count. Which, for reference, dino lora I've been posting with has dino at 3 repeats and kiato at 1 repeat(you can use whatever datasets you want, though you might want to try and make them complimentary(or contradictory) in some way).
beyond that I've also just been working on trying to tune down a batch 16(or rather, batch4grad4) config, which has been going... kind of mediocre.
>>
File: file.png (401.5 KB)
401.5 KB
Gonna post this here as well but what exactly do I select to train anima? Since it's not based on SDXL or anything, what's the meta with a 4080 dataset-wise and settings wise?
>>
I have run out of inspiration
I need to coom, but I need it to be something good
>>
>>8888146
Out of all my tests on my end UNET only achieves better prompt positioning, I do bullshit test grading on things such as top-down, birds eye view, worms eye view, angles, dark, three times per type of prompt and checkmark "if it does x, has artifacts/clean, body issue, etc." AND also without the @tag activation.

I don't know about dims honestly I won't likely understand optimization or args you lora bakers do but i'll follow it either way, so since its 32 i'll keep it as is, filesize doesn't matter to me, quality and malleability as much as possible does.
>>
>>8888154
All you have to do is enable the anima section
>>
>>8888156
pov netorase handholding, the thinking (white) man's fetish
>>
>>8888156
try this concept >>8887567
>>
>>8888156
https://exhentai.org/s/3aabd8b072/3957151-8
>>
you disgusting bobblehead pedo ill choke the funkopop out of you
>>
>p*interlyfags are shitting up my vidya too now
https://files.catbox.moe/hzbt2y.jpg
>>
https://x.com/Chief30k/status/2049937929418473803
western anime bros, we are so back
>>
Does timestep sampling matter at all? some has sigma others sigmoid, even that russel guy's confusing lora example is in a uniform type of timestep sampling.
>>
>>8888207
Garbage, looks way too 2d.
>>
>>8888211
>timestep sampling
this is needed if you want to train on some noise levels more frequently than others or just change how you sample noise levels during training
picrel shows shifted uniform and shifted logit normal distributions >>8887847
logit normal distribution focuses more on the most semantically difficult timesteps (desu that's questionable) and basically never giving the model the opportunity to learn at t=1.0 and t near 0.0
uniform overfocuses at t=1.0 and rarely learns at t near 0.0
here, t=1.0 is full noise, t=0.0 is zero noise (so it's just your image)

if you're training a character or whatever with logit normal, it may look like the model converges faster but it'll probably have funny composition/anatomy and funny little details like textures
>>
you sick bobbling child molester your head enbiggening days are numbered
>>
what is blud even going on about
>>
File: sigmoid.png (19.8 KB)
19.8 KB
>>8888238
You mean sigmoid (logit normal) is considered bad? ezscripts says that's the one recommeneded. so far looking at loras its sigmoid dominant with a mix of sigma, shift and only one I found in a uniform is that greg by russel to which I feel he hides something in his preview 3 lora.
>>
>>8888242
it isn't necessarily bad, it's just different and it has the most fucking stupid flaws that nobody actually bothered to work around (or at least i think nobody did)
use whatever you like more
>>
File: 5s.png (8.4 KB)
8.4 KB
>>8888246
I tried my best searching for more about these but considering you know more, which one is better then? I agree with you about (sigmoid) logit normal, it gives a bit of randomness on positioning/anatomy, Sigma does things better at anatomy but it has problems of the same type of positioning so i'll assume uniform is much worse.
>>
>>8888248
you should have started by trying out uniform and logit normal first and by the way "sigma", "sigmoid", and "shift" option names are beyond retarded and have nothing to do with anything. if you want to know what are the actual distributions (because this is what this option is intended to change) you'll have to look up the code.
also you can push shift value a bit higher for logit normal, say 5 or 6, it may help with anatomy
>>
>>8888257
discrete flow shift right? i'll try that out for tonights baking.
>>
https://x.com/Oshun220/status/2059619002763329652?s=20
>>
>>8888264
i assume it would be yes
>>
File: grid-noise-removed.png (2.3 MB)
2.3 MB
>>8888113
Well this took a couple hours but I removed it (with python and math and help from LLM of course) almost non-destructively. "Almost", because I think this kills the very highest frequencies so the texture looks a bit incomplete now, but that can probably be fixed by adding a little bit of actual random noise in place of the bad patterned noise back into the image.
I'll experiment tomorrow.
>>
>>8888264
also if you want to try uniform after all, keep shift lower instead, closer to 1.0, say 2.0. uniform kind of becomes not uniform because of shift. you'd still want some, but you have to find shift for each base distribution individually.

all of this is used in the first place because researchers found that some timesteps produce significantly higher loss than other timesteps on average despite not being all that difficult. that's when minsnr blown up, which is basically just lowering the loss on the timesteps which throw off training. then sd3 brainrot appeared with logit normal sampling and argued that weighting uniform loss is less efficient than selecting timesteps more intelligently than uniform. well, look how it turned out. now we're back where we started, using uniform without any weighting. some choose to pretend that the composition is alright.
>>
i cant bobble to this make it more funkopop
>>
>>8888277
i'm gonna stay with sigmoid for now, I found out bumping LR from 0.000032 upwards to 0.000042 at batch 4 made better results, but i'll still add the higher shift. (this is rex/came combo not adamw)
>>
>>8888282
>but i'll still add the higher shift
by the way, i'd expect that it'd take a bit longer to bake in wall time/epochs. with uniform it'd take even longer.
>>
>>8888286
yeah i'll slap it when I sleep later and see how it'll turn out tomorrow.
>>
unsettling lack of schizo shitposting today
>>
File: 00008-906724015.png (1.4 MB)
1.4 MB
>>8888295
>>
>>8888295
nothing to shitpost about, you are all a bunch of fucking morons but I have some more tweets to share if you want to read them
>>
>>8888124
box please!
>>
>>8888299
more anty ai thirdie tweets? no, thanks
post at least something entertaining
>>
>>8888301
>post at least something entertaining
feel free to share more for your insight about pee or scat
>>
>>8888302
*of your
but whatever, I am off to play some dotas
>>
File: ComfyUI_29652_.png (2.8 MB)
2.8 MB
Damn, posted this in the wrong general.
>>
>>8888305
nice style
>>
why is she chomping on scp dih
>>
>>8888305
this reminds me of that gorilla and donkey artist, could it be?
>>
>>8888305
that man has a weird ass bodyshape
>>
>>8888282
it's about the same speed, I set it to shift 5 and even raise the epoch from 110 to 130 ~2hours 20 mins estimate. if it goes well then i'll try shift 6 and raise it to 150 epoch including another style lora.

I've noticed on my current bake, certain complex prompts have artifacts on it while keeping the body as good as it can be, overtraining? theirs no missing fingers or hands too.
>>
>>8888307
>>8888313
slenderman looking ahh
>>
>>8888313
marfan syndrome
>>
>>
>update easyscripts
>see it going on about downloading a bunch of shit, of which includes flash attention stuff
>actually test it
>it finally fucking works and doesn't crash out
fuck yeah
>>
https://x.com/crazyclips_/status/2059303548773408796
literally me
>>
>>8888340
are you black?
>>
Noobnima when
>>
noob2edit model... whatever happened there?
>>
>>8888340
literally me except I am from the north india not norway
>>
That Time I Got Reincarnated as a Dalit in Mumbai
>>
>>8888176
that's hot
>>
>>8888305
visitor looking ahh
>>
henti...
>>
File: ComfyUI_29681_.png (1.8 MB)
1.8 MB
>>
>>8888490
my pov
>>
>>8888491
Why is your dad like this?
>>
File: 1776820213931259.jpg (498.9 KB)
498.9 KB
>>
me surrounded by 4 lucky anons
>>
1whiteboy surrounded by 4 lucky browns
>>
>>8888496
if your daughter looked like that you'd be like this too
>>
https://x.com/skulltyps/status/2059687646318702883
9/11 for bobblepedos
>>
>>8888509
+ me holding the camera
>>
File: 1759990480988164.png (477.0 KB)
477.0 KB
why are we using this jewish model again?
>>
>>8888519
>uses a ken ashcorp track for background music
fucking
lmao
>>
>>8888522
uhm, little boys are different thing, chud
>>
why can't anima do wall-eyed (e.g <_>) expression? is it stupid? no luck with natural language.catboxes are as close as i got to what i wanted.
https://litter.catbox.moe/ooer8q.png
https://litter.catbox.moe/0t9ldr.png
>>
>>8888524
what about "her pupils are looking opposite directions"
>>
>>8888526
didn't work sadly, i'm getting the same results.
>>
>>8888524
You're absolutely right
>>
>>8888315
Changing shift from 3 to 5 didint do much, possibly made the anatomy a bit wonky, it's also odd how despite having the same seed every new bake, no adjustments to the settings at all makes a entirely different type of output.
>>
>>8888532
>it's also odd how despite having the same seed every new bake, no adjustments to the settings at all makes a entirely different type of output.
you will never be able to get the same results just by keeping the same seed in sd-scripts. a lot of pytorch operations are non-deterministic.
>>
>>8888534
>a lot of pytorch operations are non-deterministic
most of them actually are. it's just that they are not necessarily used in a deterministic way.
>>
>>8888534
So it's also a bake and pray till you make.

i'll re-evaluate my own tags and see if it does something,
>>
i hate vidgens so fcking much
>>
only because nai can't do them though
>>
>>8888521
I run and train the model locally on my own computer, why should I care that SaaSjeets need to pay commercial use fees?
>>
>>8888557
because!!!!!!!!!!!!!!!!!!!!!!!!11111!!!
>>
>Ad in the op of the local lora baking thread
why?
>>
bro didn't read the row right below
>>
>>8888566
>local lora baking thread
wrong thread buddy, here's hentai generation general
>>
>>8888566
Just bake the next one yourself and remove it again.
>>
2h2g doesn't have ads just go there :^)
>>
>>8888578
link?
>>
>>
>>
nice
>>
Oh shit I think another NovelAI dump incoming, aaaaaaaaaaaaaaaaaaaaaaaahhhhhhhhhhh close your eyes bros
>>
>>8888585
are nai images really that bad if you aren't some inpainting obsessed pedo?
>>
>>8888587
Mine are pretty good and I don't have to inpaint (much). But I stole all my mixes from another bro who probably spent plenty of time discovering them.
>>
>>8888585
Imagine gatekeeping degen anime porn
>>
Whats a good combo for Adamw thats not consine? I want to try it next.
>>
File: 1774458339544226.jpg (518.3 KB)
518.3 KB
>>8888600
cosine with restarts 3 lr cycles, rex, rex with 3 lr cycles, or constant with warmup with very low lr (1e-5, 2e-5)
>>
>>8888490
>Gyaru who goes out of her way to accessorize.
>She also hits the gym very regularly and drinks her protein shakes.

I hallucinate almost as much as the AI.
>>
>>8888600
warmup stable decay if you have time
>>
my tiny pp craves more ntr
>>
>>8888606
>>8888609
thanks, i'll try consine with restarts, and then warmup stable decay later.

Yeah I do have time, if it's really long then it's a overnight thing again.
>>
File: 00065-388689477.png (2.1 MB)
2.1 MB
>>8888610
How is life in Israel?
>>
>>8888610
Best I can do is gacha slut NTS
>>
File: 1750750081954919.png (3.7 MB)
3.7 MB
>>8888617
Uploaded the wrong gen
>>
>>8888618
i lik3 the style, gimma mbox.....
.
higc..
>>
>>8888618
could have been my pov but the dude in the background ruins the vibe
>>
>>8888618
>Fat old dudes
>Known for low libido, if not erectile disfunction, and little stamina

Breaks my fucking immersion.
>>
>>8888624
yeah. shouldve been futa istead
>>
>>8888624
yeah should have been black bulls instead
>>
>>8888618
can you make the man in the doorway pakistani like me?
>>
>>8888624
yeah. should've been black futa queens instead
>>
>>8888624
we all fat old dudes with high libido itt
>>
>>8888625
Troons are not allowed on /h/
>>8888626
Bestiality is not allowed on /h/
>>
>>8888624
I can't believe my henti is doing a hecking ludonarrative dissonance
>>
File: 00002-454153465.png (1.9 MB)
1.9 MB
>Not doing at least 1 hour of exercise per day.
>>
>>8888300
https://files.catbox.moe/aw8ycj.png
Not exactly the same image since my settings are not 100% deterministic noise, but very close.
>>
Canada, being very progressive, invented treatment for fat people, both male and female. I suggest you use it. Called MAID I think.
>>
i haven't made the trip from mumbai to toronto yet but looking forward to maid treatment!
>>
why be fat when you can take ozempic and look like a ghoul
>>
>>8888638
Do it first thing after arrival!
>>
File: 00010-2160471856.png (1.6 MB)
1.6 MB
>>8888633
I'm not Neil Druckmann. I actually care for stories making at least some sense.
>>
File: 1760036085868284.jpg (510.9 KB)
510.9 KB
>>8888621
https://litter.catbox.moe/wqpg78orsxytp1os.png
>>
>>8888642
ewwwwwww he's KISSING OLD MAN CUM N SMEGMA NOOOOOOOOO!!!! AIIIEEEEEEEE BRO GOT THE FORBIDDEN DENTAL FLOSS STTTTTOOP!
>>
>>8888642
where is his nipple
>>
File: 00029-3093004920.png (979.0 KB)
979.0 KB
>>8888642
Huh. Maybe NTR/NTS is just gay domination with extra steps, after all.
>>
>>8888645
I ate it
>>
>>>8/hdg/34079
Decent gens.
>>
Lies, it's not (Dead).
>>
File: 00009-2647214693.png (3.9 MB)
3.9 MB
>>
File: ComfyUI_05859_.png (1.5 MB)
1.5 MB
On complex prompts I have low diversity on different seeds, compared to Illustrious. Any advices to increase it, besides reducing prompt and using wildcards? I don't use turbo, just in case
>>
>>8888654
Boy on the right embarrassed about size of his pp?
That's new low even for you!
>>
File: 1764276833817090.jpg (493.3 KB)
493.3 KB
>>
>>8888657
aroused, not embarrassed. the male shall be raped and the woman will watch. that is my headcanon.
>>
>>8888660
Terrifying what porn addiction does to formerly mentally well people!
>>
>>8888659
inside one of those nipples is the shrunken form of my sorry brown self. forever doomed to wander a maze of flesh
>>
File: 00002-1764433026.png (1.8 MB)
1.8 MB
>>8888662
NTA, but the buff dude is looking at the boy, not the girl. And the fucking twink blushing like a maiden makes it worse. Whatever you were trying to do here, it just looks like a scene prior to gay sex in a public bath.
>>
>>8888664
Not my gen thoooo
>>
File: 00148-1035836143.jpg (803.4 KB)
803.4 KB
I am laughing way too hard, how you morons manage to be this stupid and silly
>>
File: 1755325400819460.jpg (397.4 KB)
397.4 KB
>>
File: 2026-04-24_0082.png (2.9 MB)
2.9 MB
>>8888656
At least show what your prompt looks like. We are not in your room looking over your shoulder, you know.
>>
>>8888668
imagine being the boyfriend, admiring the amazing view and she skunk sprays your face
>>
>>8888670
>sprays your face
From her pooper?
>>
>>8888671
yes like a skunk
>>
>>
>>8888672
sir that's a honse
>>
>>8888674
*shotgun shits all over you*
>>
>>8888673
woaw
>>
File: dontlookatjuicebox.jpg (1.3 MB)
1.3 MB
when will hgg get their mascort. their mascot. it needs a mascot. agree? yes. i agree.
>>
i hold up a mirror
>>
just some bobbler and we in the full assembly today
>>
>>8888669
big boobers
paizuri from her?
>>
i don't think she's old enough for that, lil bro
>>
>pedos be like: let's slap giant balloon tits on this child, that'll trick the straights
>>
>>8888692
Personally, I prefer slapping long legs.
>>
Saars (・ิω・ิ)
>>
>>8888694
zazed. shrink me down and inject my sorry brown self into the knee cartilage of an erotic leg girl
>>
i hate tits
>>
File: 1752025921978402.jpg (404.9 KB)
404.9 KB
>>
I hate sand
>>
>>8888713
would this cure constipation or make it worse?
>>
>>
>>8888715
I hate having to work 40 hours a week
>>
consistent file name doko
>>
File: 1755832886651619.jpg (657.7 KB)
657.7 KB
>>8888716
Dunno, I'm not a doctor.
>>
>>8888724
make the anal beads come out and go in their mouths to make ouroboros anal beads
>>
Don't tell your mother
Kiss one another
Die for each other
We're cool for the summer
>>
i cast kidney stone
>>
>>8888524
I mean, you can just check the tag itself to see that the training data is pretty awful, and even then it's only 900ish images tagged with it, with most concepts that work well having 2k+ images
https://danbooru.donmai.us/posts?page=2&tags=wall-eyed
>>
>>8888730
>most concepts that work well having 2k+ images
what the fuck are you talking about
>>
>>8888717
this dudes arms are very wide
>>
>>8888731
If you go through the list of eye tags (eye concepts) you'll see that most that work have thousands of images. Very few that have less that 2k work at all.
https://danbooru.donmai.us/wiki_pages/tag_group%3Aeyes_tags
>>
>>8888737
Those aren't concept tags. But I agree that anima has some issues with tags regarding eyes and other smaller details, I dunno which part of the training caused it.
>>
>>8888737
I can literally prompt for upside down glowing swastika eyes and it'll work just fine. you just nuked the knowledge of the base model with whatever jeet slop you're using in your prompt
>>
https://x.com/friedPeeP/status/2060024446639816903?s=20
>>
>>8888741
obsessing over hating it is just as weird as liking it
>>
>>8888742
racist website, chud
>>
>>8888744
if only
>>
>>8888741
>furry
:skull:
>>
checkmate normongoloids, i like lolis with small heads and narrow hips/ lanky bodies
>>
>>8888751
you should start liking prison food in preparation for your long stay there
>>
>>8888754
this, FBI open up! XD
>>
>>8888756
unc...
>>
File: 1759385295313698.jpg (588.7 KB)
588.7 KB
>>8888725
But how would you insert the beads in the first place if it's one continuous loop?.
>>
>>8888759
it's like a large necklace. you start by shoving one end down a girl's mouth -> through her bellay -> out her anus -> into the other girl's anus -> and out her mouth. then you connect the two ends together
>>
>>8888759
box???
>>
https://litter.catbox.moe/spkpw3.png
>>
i don't like what this pov entails
>>
>>8888717
welcome back ayakon
>>
>>8888717
this nigga has to have three elbows to be able to put his arms like that
>>
>>8888618
ntrbox?
>>
>>8888791
wait nvm, thx
>>
>>8888656
What is this style? Reminds me of old Zone flashes, very cool.
>>
>>8888641
can't believe Dehya is not femdom
>>
>>8888663
Based spelunker anon.
>>
>>8888768
peter griffin if he anime
>>
>>8888805
https://danbooru.donmai.us/posts/7679379
>>
File: avr.png (6.0 KB)
6.0 KB
Is high loss a good thing per epoch?

This is set to Adamw, Consine with restarts and unlike Came, rawr/rex, its at a constant average of ~600 at the start and now at it's complete bake is ~600-460 still. Hopefully I can still continue this if its underbaked still.

Came/rex from what I recall, 340-200 from start to finish.
>>
>>8888810
means it's learning faster, no one knows if that is a good thing or not

I also want to reiterate >>8887502 because it clearly didn't stick
>>
So can you mix styles properly with Anima v1 or not? If not, the model is DOA.
>>
>>
>>8888814
define properly
it is kind of a pain
>>
File: sigma.png (87.2 KB)
87.2 KB
>>8888810
The bake looks kinda funky, i'll scrap it for now as I swapped from sigmoid to sigma as a test. likeness is there but the face is off.
>>8888813
Yeah, I did reiterate the tags, re-edit the bodies and it got much better, but not in the latest combo. I'll swap it back to sigmoid and rebake to see a difference. but once this is complete i'll attempt that long term warmup stable decay.

Kinda a bit sad, I used an old veteran's unet only/glora combo on noob here since noob 0.5 and I never even had to try out other optimizers/network, it just works. On Anima, it still works but replicates the positions way more often. (via Glora, Adamw, consine).
>>
>>8888816
>define properly
Properly is an adverb that means doing something in a correct, suitable, or satisfactory way. It implies that an action adheres to specific standards, rules, or intended outcomes

how do you do it then anon?
>>
>>8888819
Right, now define your standard, rule, or intended outcome. We could've skipped one round in you hadn't been obnoxious on purpose.

Use prompt scheduling. Either Swapping between artists or alternating at every step.
>>
>>8888820
yeah that was my bad I was feelin cheeky, but imo prompt scheduling seems like a pain, I miss the old fashioned (artist1:0.5, artist2:0.5)
>>
>>8888821
That's just how it is. Similar styles seem to mix alright more or less, but different ones just bias to one side or the other.
>>
>>8888821
>(artist1:0.5, artist2:0.5)
unc ahh prompter
>>
>[[artist1|artist2]::0.XX] [:artist3:0.XX] (3 or 4 artists is the sweet spot)
Let me guess. You """"""need"""""" more?
>>
>>8888826
I need less, actually
>>
>@fire-inc
Let me guess. You """"""need"""""" the head bigger?
>>
File: 1769911920425269.jpg (563.1 KB)
563.1 KB
>this levelp of consistency despite the difficult viewing angle

i kneel
>>
if u need more than one artist ur brown
>>
anime screencap is the white man's choice
>>
>>8888833
you will never be brown
>>
File: 1757660555599487.jpg (481.0 KB)
481.0 KB
>>
on all levels except physical i am an indian
>>
>>8888837
is curry really that good?
>>
wtf is a jordan shoe burrito gringos wtf are you doing
>>
>>8888814
just "@style1, @style 2" dude
>>
just oekaki, big huge head
>>
oekakis are /aco/ psyop
>>
>>8888814
Mixing styles with loras works very well
>>
>loras
:skull:
>>
>loras
SOUL
>>
>loras
*bobbles head*
>>
what level of obsessed does one even consider this
>>
>>8888635
thanks!
>>
>>8888659
catbox PLEASE. holy shit
>>
>>8888609
>>8888606
Whats the LR for warmup stable decay? As reference on weight I considered okay:
All are bat 4.
Came/Rex: 0.000042
Adamw/Consine: 0.00002
Adamw/Rex: 0.000038 (just baked)

Issues/nitpicks:
Came/Rex: long torso bias, fat on different bakes. (somewhat dislikes other loras of it's OWN kind but not others.)
Adamw/Consine: On spot color/body, lacks variety (same position, somehow has good lora mixing with it's own or different kind.)
Adamw/Rex: Expressive faces, cleanest so far, in between the previous two, Can enjoy lora mixing with others. I thought the bodies looked kinda funky, but what I didin't know it needed higher epochs and it's starting to look good, i'll slap it to overnight at 200 epochs, with perhaps one tick lower of it's LR.
>>
>>8888879
>Whats the LR for warmup stable decay
use constant lr.
>>
>As it stands, with these fees and the current level of Anima usage, it's still incredibly uncertain whether I will even be able to recover my own out-of-pocket training cost for Anima (over $50k) much less make enough to justify continuing to do this.
I thought comfyui paid for the training cost
>>
>>8888879
I could just revive the save state this current bake which is nice.
is adamw meant to be this long baking?
>>8888880
i'll try it out, give a good LR number for that might hit the same spot as others.
>>
>>8888883
>is adamw meant to be this long baking
adamw trains slower than came. also some people usually use 1e-4 when using adamw with a decaying scheduler
>>
_______ gacha whores
>>
https://x.com/kamidan29/status/2059981376187957407?s=20
>>
>>8888902
? that's just picture
>>
>>8888905
you don't like it?
>>
>>8888906
it's nice, but was expecting something retarded
>>
>>8888907
you are already retarded, no need to link anything external to see that
>>
anima trains really well with adamw constant at 2~e-05, but use an optimizer with kahan summation. i would also do test bakes at 512 resolution if you really want to waste time tuning hyperparameters
>>
File: ComfyUI_00234_.jpg (3.0 MB)
3.0 MB
>>8888801
>>
>>8888910
Sounds too complex for me, i'm fine with the baseline type of loras.
>>
>>8888917
dude, just use adamw8bitkahan with constant. there is nothing complex about it
>>
>>8888920
i'll test it out tomorrow, I found out adamw/rawr is very flexible even at 3k steps (150 epoch), thats tonights attempts on other style loras.
>>
>>8888924
but you already use mixed bf16 right? i don't think kahan summation will do anything significant
>>
>>8888926
I honestly have no idea, on the very first tab I have "Full BF16" ticked on, while on saving args it's save precision: FP16
>>
>>8888927
you are using full bf16
>>
>>8888928
is it a good or bad thing? i'll change it right away if it's bad but i'll do three different style loras according to the current adamw/came combo though.
>>
>>8888929
doesn't matter much. came also has kahan summation if you want to try it kahan_sum = True
>>
File: 1773441418822272.jpg (376.7 KB)
376.7 KB
>>
>>8888958
coughkneebox?
>>
File: sum.png (41.5 KB)
41.5 KB
>>8888930
didin't work, also it's waaay late at night now so i'll re-add it if you know where to tell me to place it correctly next time.
>>
>>8888970
kek update your LoRA_Easy_Training_Scripts
>>
File: 1762422946980637.jpg (426.0 KB)
426.0 KB
>>8888959
https://litter.catbox.moe/rm6r18yb7ijx63z2.png
>>
File: 00274-2478453480.png (2.0 MB)
2.0 MB
>>
https://litter.catbox.moe/orv4gpwt3m6virnh.png
https://litter.catbox.moe/ec8al9bv4u0fgr0z.png
https://litter.catbox.moe/ubqsza6796y9gr1l.png
https://litter.catbox.moe/lp9um8wtuelihptj.png
>>
how would you prompt for this kind of criss-cross top?
there doesn't appear to be a single tag for it, and prompting criss-cross halterneck will get you all kinds of halternecks with straps that cross
>>
>>8888992
criss-cross halter is still probably your best bet with heavily negging other types and emphasizing shape with NL because I had the opposite problem recently in trying to avoid it
>>
File: 1756161870826992.jpg (366.9 KB)
366.9 KB
>>8888992
"criss-cross halter bikini" works for me
>>
>>8888994
>>8888997
gotcha
>>
>>8889000
you can also make them wear other characters' outfits
>>
>>8889001
how would you prompt that? like "character name \(cosplay\) or something?
>>
>>8889010
cosplay, character name also works
sometimes the outfits might get mixed tho
>>
>>8889013
sweet, I gotta try that sometime
>>
>>8889015
not right now though
wouldn't want to post pics in an image thread
>>
>image thread
>>
File: 1769320226325054.jpg (443.7 KB)
443.7 KB
>it's called image gen
>but when you post them they are called "pictures" instead of "images"
???
>>
>image thread
he doesn't know
>>
>>8889031
i thought she had some really weirdly shaped large boobs at first
>>
File: 1763966244307038.jpg (384.8 KB)
384.8 KB
sirs why aren't you making posts of images?
>>
>>8889038
i am willing to beg for a box
>>
>>8889053
I don't get it, it looks at most like a funny meme, is this what makes /hgg/ horny nowadays?..
>>
>>8889038
nice. box pls
>>
>>8889038
this is really good. box?
>>
>>8889056
i have a big fetish for mouths
>>
>>8889056
I have a big fetish for thick cocks
>>
File: Krita.jpg (108.5 KB)
108.5 KB
>>
>>8889060
=\ OK thanks for clarification
^ (still doesn't get it)
>>
GUESS WHO'S BACK
BACK AGAIN
KRITTY IS BACK
TELL A FREN
>>
>>8889063
i'm not your girlfriend, you don't have to "get me"
>>
>>8889062
nice, highly optimized.
>>
THAT A F*CK*NG IMAGE FOR ANTS?? MOTHERF*CK*R????????!!!!!!
>>
>>8889067
don't you dare disrespect our kritten you churlish chud
>>
image optimized for average /hgg/ clitty
>>
kritten
>>
*checks time* eeyup... it's brown o'clock
>>
>kritty appears
>instant p*interlyfag seethe
lemao?
>>
kritty superiority
>>
can somebody jerk my kritty PLEASE
>>
idk 'bout jerk but i can flick it
>>
File: Krita.jpg (107.8 KB)
107.8 KB
>>
o fuck im kritting
>>
fuck need to bring my binoculars brb
>>
>>8889078
go get some ointment for your hurt asshole while you are at it
>>
notice the amazing consistency between gens? no? bring a magnifying glass.
>>
>>8889056
here is what i noticed makes people genuinely horny on /hgg/:
>petite girls that are borderline loli
>oppai loli
>the girls either wearing outfits that emphasizes their curves or focus on highlighting certain aspects of their body to which people consider attractive or simply wearing very casual loose clothes
>usage of non-/aco/ girl characters from a 90's/2000's/2010's anime or recent gacha game
>the male plapping the girl has to either be the most disgusting human being or semi attractive (but not too attractive!)
>multiple girls pleasing the same male
>acts of sex that are more than just 1girl pov vaginal
>a story can be interpreted from the image, e.g to give context to the sexy times
>interesting perspectives or/and complex backgrounds as bonus

forget anything that involves:
>milfs/mature females (unless paired with a shota, but it's still a gamble)
>handsome males
>traps/otoko no los (they have gone out of fashion)
>BBC (too much ironic posting of it has desensitized people here)
>anal
>memes
>obese/plump/chubby girls
>a style that is similar to the default look of ai anime slop
these will not get you any (You)'s or box requests.

anything i forgot to mention?
>>
>>8889083
>anal
hey anal makes me horny okay
>>
>>8889083
anal is great doebeit
>>
>>8889083
>anal
i love this
>milfs/mature females
i like this
>plump/chubby girls
i like this
>a story can be interpreted from the image, e.g to give context to the sexy times
a cool scenario i can play in my head is what makes me hard
>>
things that get me horny: ntr
things that don't get me horny: not-ntr
>>
>>8889090
based and same
>>
does blacked count as ntr
>>
File: 1776708941369494.jpg (377.3 KB)
377.3 KB
>>
>>8889094
0 story
>>
>>8889096
0 izzat
>>
>>8889094
why is he brown?
>>
>>8889097
0 laughs
>>
>>8889098
To help you self-insert
>>
>>8889094
my pov
>>
>>8889099
0 long term investments
0 net worth
7 active streaming subscriptions
>>
>>8889103
0 friends
0 inches
30 BMI
>>
can we get rid of browns?
>>
>8889094
>different pupils
>obvious ai hair intakes
>sunglasses have a weird placement/perspective.
sorry, no (You)'s will be given.
>>
30 BMI is pretty normal
>>
>>8889109
Spoken like a true american subhuman:DDD
>>
>(fat man:3.4)
>>
File: Krita.jpg (110.8 KB)
110.8 KB
>>
https://x.com/Pontifex/status/2060322763718725798?s=20
>>
it could always be worse, it could always be 256
>>
it could always be worse, it could always be 6 7
>>
>>8889113
>do not mature through relationships
Oh thank god, my artificial 1girl will stay perfect forever!
>>
>>8889112
im fuggin krittin
>>
>>8889102
https://litter.catbox.moe/6s2ofbvauikj29rb.png
>>
>>8889118
ok but img2img it now
>>
>>8889118
>ukraine flag
go black
>>
File: Krita.jpg (425.3 KB)
425.3 KB
>>
>>8889118
funny but are you too fat to lie on your stomach or something
>>
AIIIIIIEEEEEEE!
>>
>>8889118
if i could draw this good id never use AI
>>
>>8889124
Yes, I'm too fat to lie on my stomach but I have no issues holding a handstand for several minutes when it's a bvll having its way with my waifu
>>
>>8889118
He is obviously lying in the same way as 1girl.
Also you made 1girl erotic and not an ugly congregation of fat so it's not even the same picture anyway.
>>
>>8889094
if she started beatboxing hard enough with her butthole she could make the bvll's penis bob up and down with the tune
>>
>>8889127
based handstand enthusiast
>>
>>8889123
o god my kritty!!
>>
anima autism mix when
>>
File: guinea.png (59.1 KB)
59.1 KB
>>8888973
>fix: REX, remove reduction of total steps
>update 2 hours prior before recommendation and update
hope it does something.
>>
>>8889139
>fix: REX, remove reduction of total steps
that's for validation loss
>>
is there any good 2d anima finetune yet
>>
this but 3d
>>
this but anime
>>
this but jav
>>
>>8889142
poop
>>8889143
kaka
>>8889144
feces
>>8889147
shit
>>
why is bro listing Indian delicacies
>>
all my niggas krittens nigga heil kritty
>>
https://www.youtube.com/watch?v=m57LQjRbhcA

this but krita
>>
https://x.com/kianamaiart/status/2060290961171980533?s=20
>>
only black artists should be allowed to draw BLACKED henti
>>
what about BROwns?
>>
>>8888879
>Whats the LR for warmup stable decay?
I use 3e-6 but that's for finetuning, you can probably double it for loras.
>>
>>8889165
shouldn't those losers be on that one shitty "progressive" twitter clone that's become so irrelevant I can't even remember the name of it anymore instead of heil hitler xitter?
>>
File: 1766697567570560.png (3.0 MB)
3.0 MB
>>8889118
>>
>>8889172
holy based
>>
>>8889172
my pov
>>
>>8889169
will try it later, thanks!

Also whats this [tag|tag2] type of prompt? that's new to me, a true 50/50 mix?

And mass weighting i'm used to ( tag, artist, etc:1.2) type of mass weights. rather than (tag:1.1 ,artist:1.15, etc:1.3), was I wrong all along? the prompts look fine on my method since noob, but could be anima new.
>>
>>8889177
>[tag|tag2]
there's one syntax that will apply the tags on alternating steps and another syntax that will first apply tag1 for x % of the prompt and then tag2 for y %, but i can never remember which one is which.
>>
>>8889172
based and still my pov
>>
>>8889170
bluesky is so dead there isn't anyone to preach to
>>
File: 1759457399832222.jpg (264.5 KB)
264.5 KB
Are you telling me that BLACKED 2hoes are hurting the BIPoC artist community on X (formerly Twitter)?
>>
>>8889165
someone ask them why they have "aiart" in their name if they are anti ai
>>
>>8889193
shut it down
>>
>>8889172
kino
>>
File: 00009-3867342986.png (1.7 MB)
1.7 MB
>>
File: Krita.jpg (106.6 KB)
106.6 KB
>>
>>8889222
By the age Sasha-chan could get that kind of body, she probably would have already died in a freak accident involving an hydraulic press and several dildos modeled on the most famous race horses of Japan.
>>
>>8889223
o im krittenin
>>
File: 00005-2213372260.png (1.3 MB)
1.3 MB
>>8889224
she'll be a menace to all shotas
>>
sasha-chan would never fuck sh*ta
>>
I would fuck a shota
>>
File: 00003-2213372260.png (1.8 MB)
1.8 MB
>>8889231
>>
>>8889230
That's like half of haguhagu manga with her tho??
>>
>>8889233
I thought the point was that she never fucks the shota and only has sex with older men
>>
why are shitafags like this?
>>
any cock size comparison?
>>
>>8889178
>another syntax that will first apply tag1 for x % of the prompt and then tag2 for y %
it's just
[tag1:tag2:x.x]
>>
how do i gen 17 cm penis
>>
>>8889241
Just "penis", it's average.
>>
(๑´ڡ`๑)
>>
>>8889246
abc
>>
>>8889232
Why is her head so wide?
>>
>>8889228
blud has a gaping void for a belly button
>>
>>8889248
it's pedo artist
>>
File: 1751407546690.png (2.0 MB)
2.0 MB
>>8889230
this, even in the "vanilla" manga she loves ugly fat bastards and is canonically not a virgin.
>>
>>8889255
>/hdg/ designs a male character.png
>>
>>8889224
>>8889228
Isn't the author of this thing some edgelord who keeps drawing this girl getting strangled, beaten, raped, drugged and so on? I doubt she willever reach 14.

Although I heard he made completely SFW comic. I just don't know anymore.
>>
File: 00003-2213372260.png (1.9 MB)
1.9 MB
>>8889257
she'll reach whatever age ai wills her to reach
>>
>>8889257
He always did some SFW on the side like exhentai.org/g/775203/5e87a59fa9/ and he doesn't just draw hardcore abuse stuff.
Sasha's books are mild compared to some of his nastier work, probably because she is legitimately his waifu. Also, he actually did a book where the shota gets to fuck her https://exhentai.org/g/3546198/a37dad9926/
>>
I don't get it, you can just search haguhagu on exhentai and see like 15 doujins where sasha fucks shotas, what's with the cope???
>>
>>8889262
Not canon
>>
what's with all the cultural appropriation, yikes
>>
>>8889260
give me the metadata so i can generate her scatting on the shota
>>
>run 1000W pc to train loras
>run 3600W air conditioner to keep room cool
>all so I can fap to machine hallucinations of gacha whores
>>
>>8889267
How many kW to train the Nangong Yu lora
>>
Best pedo artist?
>>
>>8889269
for me it's gomennasai
>>
>>8889267
It's a life.
>>
>>8889269
firedotinc
>>
we are so back https://civitai.red/models/2660555/anima-wildcrad-anima-simple-sex-01?modelVersionId=2987475
>>
File: tard-kun.png (695.2 KB)
695.2 KB
>>8888524
I got this image completely unprompted a few days ago. It would seem it knows. I can't check your links cause you're retarded and used litter but have you tried upping the strength? Depending on the tag you can go upwards of :4.0 without frying anything.
>>
Does the easy training scripts sd-scripts backend not start automatically on linux' run.sh or did I just fuck up massively?
>>
>>8889278
>have you tried upping the strength
nta but i tried (wall-eyed:5) and it still didn't work
>>
>>8889267
my machine probably uses less than 350~w to actually gen things(absolute peak is probably under 450w but that's only under heavy CPU load).
I have a window AC unit that's less than 500w
unironically I use more electricity in the winter keeping the room warm than I do in the summer to keep it cool.
>>
>>8889279
"run_local": true in config.json, and set backend url in the ui to http://localhost:8000
>>
File: 00010-2594584973.png (1.9 MB)
1.9 MB
>>8889265
https://files.catbox.moe/aszf6y.png
>>
File: joy.png (92.3 KB)
92.3 KB
Some vibe coded forge neo extensions I've been using for myself, posted a couple here before but they have been iterated on it since. All have long readme's included, integrate with each other and other extensions (i.e. tag autocomplete) and so on..

https://files.catbox.moe/bhy4n1.zip >implementation of https://github.com/KeithZ117/Comfyui-anima-sampler in forge neo, also added presets for img2img upscaling passes/etc.. I recommend "Cosmos 2.5 Default — UniPC · RF linear shift5" with 35 steps between 5-7 CFG and an 0.5 denoise send to img2img upscale (no img2img upscaler selected) with "[Upscale] Detail — PC3 · S-tail shift5 · clean pass" using 6 CFG.

https://files.catbox.moe/02ok2g.zip -
>Resolution picker with a shit ton of options. standard preset total ≤ 2176; high-res preset total 2240–3072 (capped 2048 per side); on send to img2img/inpaint, Resize-to is auto-set to a multiple-of-64 target whose W+H lands in 2816–3200. Every Generate click silently logs the current positive prompt into a 100-entry session ring (prompt log)

https://files.catbox.moe/05ykxs.zip
>Adds a new tab for very in-depth Anima prompt building/formatting, also has a booru tag grabber more advanced than other extensions (i.e. ranbooru) that supports danbooru/gelbooru/aibooru/rule34 and integrates with prompt builder/other webui pages(extensions). Has advanced filtering to only pull artist tags between a certain post count, blacklist words, etc, etc, etc...

https://files.catbox.moe/i4gjbl.zip
>Implements a new tab for JoyCaption, will auto download the desired models and has integration with other tabs/extensions. Includes a lot, will also auto downscale images to around 1024x1024 if images are sent from a booru grab or something, its pretty autistic.
>>
virus
>>
>>8889283
Yeah they were both set to local, but for some reason just running run.sh in the root folder doesn't start the backend automatically too as it did in windows. Have to manually start it. Anyway, now that I can start training, what are the recommended settings for Anima anyway? Do you need to set some noise offset like Illustrious recommended or anything?
>>
>>8889292
>Do you need to set some noise offset like Illustrious recommended
never use noise offset. you can enable pyramid noise for anima but it doesn't seem to do anything for me so i don't use it
>>
>>8889289
joycaption has been powercrept by qwenVL
https://old.reddit.com/r/StableDiffusion/comments/1jh8b3k/update_qwen25vlcaptionerrelaxed_opensource_image/
>>
File: 1768223708703113.png (1.3 MB)
1.3 MB
>>8889246
good morning sirs
>>
>>8889294
Do you add all three models to the anima args section or just anima base and the vae? Is training the unet only more effective for anima? Not sure where to start desu
>>
>>8889306
add all three models to the anima args section
>Is training the unet only more effective for anima?
i prefer training only the unet. but you can also train the te
my config https://litter.catbox.moe/yg7ez5yxub7scau1.toml
lr 3e-5 to 4e-5, around 1000 - 1600 steps at batch size 2
>>
>>8889303
Your uncensored tag, sir?
>>
File: 00006-4230441355.png (1.4 MB)
1.4 MB
>>8889311
how do you know he doesn't just enjoy mosaics
>>
>>8889313
nice style
>>
>>8889315
anime if vector
>>
>>8889317
like the guy from despicable me?
>>
>>8889309
not another toml for me to test...

>rawr/came

oh its the same one, is low dim a good or bad thing? i'm always 32/32
>>
File: 00050-3171053536.png (1.6 MB)
1.6 MB
>>
File: Final Scene 4.jpg (1.7 MB)
1.7 MB
>>8889303
Hello.

And for those not in other thread, I've made a Louise succubus corruption set. More here >>>/d/11521980
>>
anime if Western
>>
>>8889270
Why did he completely stop making ecchi stuff? I feel like something happened that made him only do sfw stuff from then on.
>>
Using char loras significantly reduces diversity across different seeds. Not sure if this can be fixed, need to train myself to test. On shitmixes diversity is even worse

https://litter.catbox.moe/5rn15wjpynqhefl1.png
https://litter.catbox.moe/eeocwj2x5q149xxq.png
https://litter.catbox.moe/tabwrjqnc1pjk80y.png
https://litter.catbox.moe/04i5tdljz3x0m02a.png
>>
>>8889330
your Louise looks aco and adultish...
>>
>>8889336
Possible problem

https://www.reddit.com/r/StableDiffusion/comments/1tobzgq/comment/oo03b0f/
>>
>using alphonse for adult female
yikes
>>
>>8889336
why are you using a jeetvitai lora for a character with 1.4k dabooru posts
>cumrag
Oh that explains it
>>
>>8889336
Personally I play with lora weights way more on Anima than I ever did with any previous model even though I used to be a staunch believer in that it has to function perfectly at 1.0 or it is trash.
>>
>>8889336
>overtrained civitjeet lora is overtrained
shocker
>>8889343
kek you beat me to it
>>
>>8889343
>why are you using a jeetvitai lora for a character with 1.4k dabooru posts
Obviously because I need exact same screencap style, not a doujinshi based on it
>>
>>8889346
>I need exact same screencap style
Then why are you using a CHARACTER lora you fucking retard
>>
>>8889336
Always the case with loras. Some conceited anons will tell you that THEIR loras don't do that, only civit loras do. ALL loras do that. Can't train a lora and not make the model noticeably worse in everything else.
>>
>>8889347
>Then why are you using a CHARACTER lora you fucking retard
Don't you realize that style includes specific facial features of a character, imbecile?
>>
>>8889347
This guy has never used nochekaiser's loras before
>>
@chat I think I'm getting trolled
>>
>>8889348
It wasn't like that on the Illustrious/pony, there was always good diversity on different seeds
>>
File: 00003-1538829855.png (1.3 MB)
1.3 MB
https://files.catbox.moe/rc4owx.png
>>8889353
no way what gave it away?
>>
>>8889352
I'm not well versed enough in jeet-jitsu to know who that is.
>>
File: 1764053574109363.jpg (394.1 KB)
394.1 KB
>>
catastrophic forgetting is not a joke, Jim!
>>
>>8889355
my mother
>>
>>8889354
There is a practical solution and it is lora scheduling. Good luck though, I'm not sure it was ever possible with forge and for some reason there is a bug with anima that makes it impossible in comfy too.
>>
>>8889366
the real practical solution is just training a lora that isn't shit
>>
>lora
>not shit
Heh
>>
NovelAI doesn't have this problem
>>
The only thing worse than character lora is maybe pancreatic cancer, and that's probably a tie.
>>
>>8889366
>lora scheduling
Thanks, sounds interesting, I'll try to vibecode something
>>
>cumrag retard is going to vibecode some useless shit instead of just solving the actual problem
lol
>>
>>8889376
That's the cumrag way, never try to find a reasonable solution just add more NOODES
>>
>solving the actual problem
So training an edit model? OK have fun!
>>
File: 00007-1761991756.png (1.7 MB)
1.7 MB
what my mother thinks will happen when I leave plates with crumbs on the window sill
>>
>>8889376
To solve "actual problem" I'll need your address, but I doubt you'll give it to me, so I'll have to improvise some workarounds
>>
edit models are snake oil
t. jeetvit's top loraslopper
>>
who gets this upset over loras existing
>>
>obsessed with some oldtroon anime
>cumrag
>Euler A
>too retarded to use a lora that wasn't trained by a Manila resident or just using
>bona fide Internet Though Guy
Is this the first /hgg/ native pillar
>>
>>8889381
@grok which Brazilian superstition is this post making a reference to?
>>
>>8889381
foul smelling creature
>>
https://x.com/AziEmblemFates/status/2060410871290229109
is this heckin dorontabi reincarnation?
>>
>just realized forgekeks never had lora scheduling, something that I've been using for years and took for granted like artist scheduling
Holy shit stone age fuckers.
>>
>>8889386
there is an infinite amount of dunning kruger cumrag retards
>>
>lora scheduling extension exists since 2023
>>
>>8889395
loractl for a1111? lol that doesn't work since 2023
>>
does this gen make you guys horny
i tried my best!:
https://litter.catbox.moe/yxrfq0.jpg
>>
>>8889398
give us an xray view of a dildo in his anus and it would be perfect.
>>
>rule 34
>nostrils on female
vomit
>>
ok pedo
>>
>>8889398
shota with face and bdsm on male are kinda gay
>>
It's cockroach post, why are you surprised it's gay anyway =\
>>
>>8889398
really nice, shotabox?
>>
what happened to that guy advertising the ai influencer thread on /g/
>>
>>8889398
they should tape my shrunken, sorry brown self onto the shota's penis before paizuring it, only for the tape to eventually come off. leaving me trapped and forgotten in a hot and humid hell of tit-flesh and cum (which i drown in).
>>
>>8889398
No. ss, femdom, nor that princess are my things.
>>
>>8889409
Photo of your hand or I will always think that these posts are typed by snow white anon from norway.
>>
you know, being white isn't all sunshine and rainbows, sometimes I get sunburn
>>
>>8889411
you don't interest me
>>
>>8889412
all whites are rainbows
>>
https://litter.catbox.moe/85p5avcj4mpir2p4.png
>>
>>8889399
>>8889400
ok! i hope this makes you guys even more horny. i added the dildo and removed the nostrils:
https://litter.catbox.moe/rie8gz.jpg
>>
>>8889398
>>8889418
sirs, box pleaze
>>
>>8889418
>still a character from non-porn media
>but now with faggot shit (male_penetrated)
1 step forward, 10 steps back.
>>
>>8889421
>1 step forward, 10 steps back.
more exercise than you'll ever get, fatass.
>>
>>8889406
>>8889420
https://litter.catbox.moe/rvcqji.png
There is some schizo prompting because i was trying out things, and then inpainted in the rest. But the catbox is the base gen.

kind regards,
mipmup/gforceanon :))))
>>
>>8889425
mr. sir, please make g-force lora for anima
>>
>>8889427
nah
>>
>>8889418
could you add back the shota's nostrils and make the girl lick his boogers? also add in his prostate. that would make this image truly perfect :)
>>
File: decay.png (37.2 KB)
37.2 KB
>>8888930
Kahan with came/rex was so-so, about the same, could be better but I removed/added new tags, edit a few pics.

i'm attempting warmup stable decay right now and see if its good, better or not. After that, I might have to ask someone how to properly set up kahan_sum on adamw/constant next, unless it's just the same as came with kahan_sum = true
>>
>>8889439
kahan is just a speed up/memory saver. if you want top notch quality you shouldn't use it
>>
Does bucket resolution affect anything for body format? It's set at 256 min and 3072 max, for noob where all the bodies are good and okay, it was set to 2048 from the old kohya gui.
>>
>>8889445
Doesn't matter. you'll likely never have any images in your dataset that make use of these buckets at the edge of the distribution
>>
File: constant.png (27.3 KB)
27.3 KB
>>8889449
thanks, this is the next bake test once stable is complete. (no khanon), this LR rate just feels similar to glora's adamw/consine combo but i'll try anyways.
>>
>>8889439
A LR rate that extremely low barely made any progress at 1980 steps.
>>
File: 163450_00006_.png (3.3 MB)
3.3 MB
>>
https://files.catbox.moe/5yrydd.png
>>
>>8889498
the /hgg/est gen i've ever seen
>>
ANON
>>8880038
>>8880038
>>8880038

GIVE ME A CATBOX AND MY LIFE IS YOURS!
>>
>>8889239
I did this one the other day
>https://files.catbox.moe/x5s6n0.png
>>
>>8889504
You must be retarded.
>>
File: .png (139.5 KB)
139.5 KB
>>8889504
>>
>>8889505
unironically least gay gen in hours
>>
>>8889507
how the fuck?
>>
>>8889510
script from the OP https://rentry.org/hdgcb
>>
File: 00239-2953476748.png (835.2 KB)
835.2 KB
ah yes. shitty 00's era web art aesthetic achieved.
>>
>>8889513
that's fucking twin_(tt_lsh) with a very bad pixel filter over it
>>
File: 00249-2953476748.png (952.2 KB)
952.2 KB
>>8889514
actually it's a shitty lora trained for 20 minutes on this single, shitty image
https://files.catbox.moe/9z6jga.jpg
>>
>>8889515
is this actually 00s or just contemporary le quirky shit posing as such
>>
File: 00255-2953476748.png (957.2 KB)
957.2 KB
>>8889518
contemporary le quirky shitposting
I don't even actually know the source artist because apparently this is the only image of theirs I saved and I didn't fully write out their actual xitter username(oops).
>>
>>8889520
https://x.com/made_in_sonohen/media
>>
File: 00268-2953476747.png (1.2 MB)
1.2 MB
>>8889521
damn, you're good. also now I understand why I only saved that one image and why the only label I had for it was sonohen(I had no faith in the artist doing more of what I wanted of them).
anyway if anyone wants this piece of shit 4mb lora mediafire.com/folder/i3ekeugv3zx3u
activation tag @j34
style on its own is kind of shit but it mixes on top of artist calls kind of fairly okay? depending on what they are, anyway.
>>
>>8889526
just img search and google found the instagram account, although it's really shit lately
>>
>>8889526
Pretty cute, this looks useful for a style I want to go but only draws one type of girl of the same everything, always in coats but so much enviroment and theme.
>>
>>8889450
0.00002 weighting on this adamw, constant combo didint turn out as good, honestly outside of adamw/consine glora, anything baked in adamw really loves to turn things as if it's made by hogarth, always tall, shiny, etc, torpedo breats, etc

Back to came testing.
>>
>>8889538
https://github.com/rmatif/ai-toolkit/tree/anima
could you try ai-toolkit? the sd-scripts implementation was vibe coded by a retard so i don't trust it much
>>
File: 214435_00003_.png (3.1 MB)
3.1 MB
>>
>>8889566
nope, if oldies here say ez scripts i'll stay ez scripts... unless kohya gui comes back which is never ever, simply because I can always ask questions, get recommendations on baking, format issues and load tomls given.
>>
>>8889570
wtf is this perspective also raiden is for big hilichurl cock only, many 3d animations can confirm this
>>
https://x.com/QQ_Tiny/status/2060437898793553938
please, go check on pubmed whether the eyes are of age
>>
>>8887847
>also note that shift is the only "correct" way to adjust sampling schedule of flow matching models, "beta", "karras", "exponential", "ays" or whatever were not made for flow matching and when you use them there's a ton of math fuckery going on without a good reason
Then what should you put in scheduler?
>>
File: 233450_00008_.png (3.9 MB)
3.9 MB
>>
>>8889585
nice, very wai v19 like
>>
>>8889586
Yes and?
>>
>>8889587
stroking my clitty to this masterpiece
>>
>>8889585
nice. more pls
>>
>>8889588
Based Klitty
>>
For REX, is restart 3 optimum or will higher/lower hinder it in some way?
>>
Does chat have any preference between konan or nochekaiser? Both have loras of a character I need but which one is known for high quality?
>>
>>8889596
both. both are good
>>
File: 002210_00002_.png (3.5 MB)
3.5 MB
>>
https://x.com/Maziodyne_/status/2060095102442504616?s=20
>>
File: 1749099445929388.jpg (451.3 KB)
451.3 KB
wher all my lil brown bros @
>>
File: 00027-1990000361.jpg (321.2 KB)
321.2 KB
>>8889627
wtf I did a gen with almost that same setting back on pony days
stop hacking my computer
>>
Very original
>>
Are you telling me that more than one person ever genned an ass-focused pov reverse cowgirl picture? That's fucking crazy man.
>>
are you going to stop hacking my computer tho
>>
>>8889627
brown & bloody proud!
>>
/hgg/ is for penis focus
>>
/hgg/ needs to center the lived experience of black bodies
>>
>>8889189
>>8889313
>>8889417
box please
>>
File: 1759712506338891.jpg (420.0 KB)
420.0 KB
>>8889628
Sorry, I already have Ring 0 access to your system and I am now downloading all your prompts.
>>
>>8888920
that one implementation of adamw8bit with kahan that's going around is not adamw, it's using adafactor's update clipping. if you liked it you should try adafactor or stable adamw
>>
>>8889582
if you want to avoid math fuckery? you want to use diffusers because comfy sampling backend will fuck with math regardless of what you put in it.
if you just want to see to see the native schedule you should use "simple"
>>
>>8889650
https://github.com/67372a/LoRA_Easy_Training_scripts_Backend/blob/02bb6afaceea609d736412cd3f02c21b64d3b0e5/custom_scheduler/LoraEasyCustomOptimizer/adam.py
there won't be update clipping if stabilize=False right?
>>
>>8889648
I wish I were a little smug bro instead of a little brown bro :(
>>
>>8889653
that would be correct.
>>
>>8888888
>>
File: 1752644893987713.jpg (421.9 KB)
421.9 KB
>>8889639
https://litter.catbox.moe/92nrbp5scfzi5zh1.png
>>
>>8889648
flatulence?
>>
>>8889671
no e621 = no braps
>>
>>8889673
there is no heaven
>>
>>8889674
just... train a lora
>>
e621 contributed useful learning in terms of anatomy and concepts and I'm tired of pretending otherwise.
>b-but it ruined my pure animu aesthetic with le sepia and high contrast!
partly skill issue on part of the trainers and mostly SDXL being a dogshit architecture that wasn't completely unusable out of pure luck
>>
this, real e621 has never been tried
>>
https://litter.catbox.moe/vi281by94ti3koil.png
>>
>>8889680
more like has never been FRIED ;^)
>>
File: file.png (1.3 MB)
1.3 MB
Anima style lora recs?

>>8889666
>[@mikoyan, @honjou raita, @haku89: @linhangwei, @dishwasher1910:0.333],
NTA but is this actually stable? I've tried it a bit before with limited success, but I'm not sure if it's the specific artist combo's fault.
>>
>>8889682
there are little germs and cheese particles floating inside that pee. and what are those germs n cheese made out of you ask? me. that's right. my sorry brown self has been shrunken down into microscopic levels before being forced to undergo endless mitosis. my little peeples now have no choice but to colonize your little penis for survival
>>
>>8889685
I don't know either, I was fucking around with mixing and made that gen mostly as a shitpost.
>>
>>8889686
Based molecular anon
>>
>>8889633
>>8889635
in the name of BBC, gaymen.
>>
>>8889652
>because comfy sampling backend will fuck with math
I'm on forge neo though, is it the same?
Regardless I'll try those
>>
>>8889678
You could always just find a very representative artist of e621 and add it to the negatives
>>
https://files.catbox.moe/omgh4b.png
>>
>>8889652
Thoughts on this videcoded solution? https://github.com/KeithZ117/Comfyui-anima-sampler/tree/main
>>
head way too small
>>
what's with comfytroons superseding pedos and naifags as the most annoying posters itt as of late?
>>
https://github.com/Wenaka2004/comfyui-anima-ipadapter
owo wats dis
>>
>>8889720
https://huggingface.co/Wenaka/anima-ip-adapter-dataset
no weight yet
>>
>>8889722
soon hopefully
>>
just
>>
Never hopefully, stop undermining NovelAI superiority!!
>>
>>
>>8889727
Erotic upper body.
>>
>>8889727
erotic cum on one male boob. box?
>>
>>8889720
>>8889722
Is it going to be unusable abysmal dogshit like all IP-Adapters before it or will it be actually decent like NAI's Character Reference?
>>
>>8889730
>decent like NAI's Character Reference?
lol
>>
File: 1748790318587855.png (2.7 MB)
2.7 MB
>>8889731
I mean it's no edit model but it sure as shit beats the inpaint trick.
>>
>>8889730
yes
>>
>>8889685
>but I'm not sure if it's the specific artist combo's fault
Yeah prompt scheduling can really go haywire with certain artists mixes. No real explanation why but if a certain combination is giving you something low quality half of your gens, then swap out an artist, change the timing of the steps, etc. I like to end a schizo mix with an artist with a predominantly harsher style if the early steps are soft or vice versa.
>>
>>8889727
buddingotokobox?
>>
File: 1760707033486837.jpg (433.4 KB)
433.4 KB
>>
>>8889736
Me on the left
>>
https://files.catbox.moe/lixymy.png
>>
you guys really seem to know your stuff with how the funny magic box generates its images. good for you
>>
>>8889737
nice suit
>>
>>8889702
>I'm on forge neo though, is it the same?
yes
>>8889716
slop
anon, you can pack the entire flow matching sampling into like 5 lines

timesteps = torch.linspace(1, 0, num_steps + 1)
timesteps = shift * timesteps / (1 + (shift - 1) * timesteps)
for t_curr, t_prev in zip(timesteps[:-1], timesteps[1:]):
pred = model(timesteps=t_curr, ...)
img = img + (t_prev - t_curr) * pred
>>
File: 1751326977660014.png (1.0 MB)
1.0 MB
>>8889732
what am I supposed to be seeing here?
>>
>>8889743
It somewhat looks like the character in the reference image. just like nai somewhat looks like anime.
>>
>>8889743
NAI Character Reference vs the inpaint trick with Anima + LLLite ControlNet.
Also your pic has a very blatant error - she's supposed to have a blade sticking out of her forehead, not an ordinary flesh/keratin oni horn. NAI got it right like 75% of the time, Anima only maybe 15%.
>>
File: 00080-992143299.png (1.1 MB)
1.1 MB
>>
>>8889736
javactorbox?
>>
File: 1750862063907095.png (1.0 MB)
1.0 MB
>>8889748
>she's supposed to have a blade sticking out of her forehead
Oh sorry I couldn't tell that from the tiny portrait in the top right and your gens looked like a black horn. Still, I wouldn't bother installing your snake oil.
>>
>>8889742
Yes, you're right this github is overfilled with AI engineered slopcode but it does implement the flow matching sampling as far as I can tell.
>>
just train a lora bro
>>
>>8889759
>but it does implement the flow matching sampling
so..? comfy also implements flow matching in its core
>>
File: 02560-1135340354.jpg (2.0 MB)
2.0 MB
>>
>>8889772
estrogennokobox?
>>
>>8889774
It's called a girl, lil' fag...
>>
File: 1776444644613410.jpg (542.9 KB)
542.9 KB
>>
>>8889776
Let's see what the box has to say.
>>
Hand it over, that thing, your onee x shota gen
>>
this but oni x shota
>>
>>8889781
as long as the oni is a busty girl
>>
>AssertionError: extension access disabled because of command line flags
What the fuck? I have nothing like that set
>>
>>8889776
>>8889778
time so see what the box says:
>>8889774
https://litter.catbox.moe/3gf7fb.png
>>
>(manboobs:2)
lol
>>
>draw girl
>call it a boy
ai edition
>>
File: 00156-2263549779.png (1.8 MB)
1.8 MB
>>
File: 1771952281984190.jpg (398.1 KB)
398.1 KB
>>
>>8889790
me on the phone
>>
>>8889627
thats a kid
>>
>>8889796
she's just a petite girl
>>
>>8889786
>time so see what the box says
Yeah but not the same picture, box doesn't have booba.
>>
Juat report it and let the janny decide
>>
>>8889727
>post NovelAI gen
>call it anima-base-v1.0.safetensors
Blatant false advertisement for local.
>>
File: 1758718983045799.jpg (414.4 KB)
414.4 KB
>>
https://files.catbox.moe/hs4r77.png
>>
>>8889729
>>8889735
https://files.catbox.moe/0hm5cr.png
I'm playing around with the new slop sampler, also ran out of female characters to gen so figured it was worth trying the genderswap tag.
>>
>>8889786
>po-ju
Dangerously Based.
>>
File: 00198-1215814227.png (1.7 MB)
1.7 MB
>>
File: 02577-2363663582.jpg (1.8 MB)
1.8 MB
>>8889790
>>8889820
phonebox? i unironically love your images.
>>
Why does anima sometimes just default to flat color regardless of artist or style tags
>>
>>8889736
puff poosy.... squeezee.... thats da juice they b drinkin box?
>>
>>8889822
I unironically think your images are vomit-inducing and so is every post of yours.
>>
>>8889823
now you know what you should put in your negs
>>
File: 00210-3380344553.png (1.8 MB)
1.8 MB
>>8889822
it's a very distinctly adequate level of shitty. I can't really describe it, it scratches a dumb itch, though.
https://files.catbox.moe/1wio50.png
https://files.catbox.moe/zo0rhu.png
mediafire.com/folder/3adz8dzglpqpi
the one aliasing is the sonohen lora, the uenomigi is doing some styling/color stuff.
>>
https://x.com/JPT_Struggles/status/2060518087758323935?s=20
>>
>>8889826
>implying it wasn't in the negs already
be a retard somewhere else
>>
literally me when using the slopmachine
>https://x.com/iwatanishi0726/status/2060531361371598978?s=20
>>
>>8889823
maybe mix the base model with a 2.5d shitmix
>>
File: 00214-3380344553.png (1.8 MB)
1.8 MB
>>8889827
also for reference of just what the uenomigi lora is doing, same seed/prompt but without the uenomigi lora
>>
File: 00234-4172700711.png (1.1 MB)
1.1 MB
>>8889833
that said, you can just drop the sonohen lora on top of pretty much any other artist lora, adjust the artist lora weight down a little and it should work fairly okay? probably works a lot better for artists with flat colors/that aren't super heavily detail oriented, but I haven't really tested it too heavily.
https://files.catbox.moe/9hl2pc.png
>>
>>8889842
post a close up image of your erotic, gaping mouth so i can fry a lora on it
>>
what's with the obsession of pushing for the shittiest style possible
>>
but enough about anime screencap fags
>>
it's really satisfying to push out a big, fat log
>>
I somtimes gen loli but ultimately I always end up wondering: what would this look like if she had fat sagging hangers?
>>
imagine genning
>>
File: 02589-398112648.png (1004.0 KB)
1004.0 KB
>>8889827
>>8889833
>>8889842
thanks! your 1dim loras are amazing. and thanks for the other tips, i might use them in future gens as part of a mix.
>>
i don't have to imagine, I do it every day as well as any other of my hobbies
>>
I need to gen
so more electricity is consumed
so more oil is needed
so the middle east gets destabilized
so more brown people die
>>
>i might use them in future gens as part of a mix
rip
>>
>>8889856
they come to your house, not die. lil bro
>>
shitposting on a dead bangladeshi basket weaving forum 24/7 is not a hobby
>>
WAIT! Hold it!!! Are you guys telling me that... you are SUFFERING????
SUFFER!!!!
>>
im into that, lil bro
>>
>>8887735
>>8887946
>>8887948
>>8888113
>>8889666
these images make me horni. thanks for doing good gens.
>>
File: 1762487889881391.jpg (329.9 KB)
329.9 KB
>>8889863
Getting horny from other anons' gens is gay.
>>
>>8889863
you too
>>
Has anyone had any success prompting for three different girls with distinct features and keeping features from blending together?
I am trying to do this
>the girl on the left: \(dark blue hair, pale skin, curtained hair, medium hair, parted bangs, purple eyes, head wings, bat wings, brown wings, harem outfit, revealing clothes, purple breast curtains, see-through clothes, puffy nipples, purple see-through mouth veil, purple bridal gauntlets, grabbing own breast, deep skin, large breasts, looking at viewer, puckered lips, narrowed eyes, naughty face \).
>the girl on the right: \(dark red hair, double-parted bangs, blunt bangs, blunt ends, long hair, skin-covered horns, pointy ears, orange skin, revealing clothes, criss-cross halter bikini, red and white striped bikini, skindentation, covered erect nipples, mini wings, dragon wings, large breasts, looking at viewer, fang, smirk, open mouth, v-shaped eyebrows, smug \).
>the girl in the middle: \(light green hair, twintails, double bun, beige eastern dragon horns, short hair, flipped hair, hair between eyes, detached ahoge, pink eyes, green eastern dragon tail, white bikini, fur-trimmed bikini, large breasts, licking penis, looking at viewer, head tilt, half-closed eyes, raised eyebrows\)
>>
>>8889878
i would just use character tags without outfit and appearance tags
>>
>>8889878
I've also tried straight up NL
>and I've also tried straight up NL
>the girl on the left has dark blue hair and curtained medium hair with parted bangs pale skin. she has purple eyes and bat head wings as she is wearing a revealing harem outfit compromised of purple see-through breast curtains with visible puffy nipples. she is wearing a purple see-through mouth veil and purple bridal gauntlets. she grabbing her own breasts resulting in deep skin. she has large breasts and is looking at viewer with puckered lipsa and narrowed eyes with a naughty face,
>the girl on the right has dark long red hair with double-parted bangs and blunt bangs with blunt ends. on her head are skin-covered horns and pointy ears. her skin is orange as she is a demon girl. she is wearing revealing clothes such as a criss-cross halter bikini with red and white stripes featuring heavy skindentation and covered erect nipples. on her back are mini dragon wings. she has large breasts and she is looking at viewer with a smug smirk and v-shaped eyebrows and an open mouth with her fang visible,
>the girl in the middle has light green short flipped hair in twintails and double buns with beige eastern dragon horns on her head with hair between her eyes. a detached ahoge is visible on her head. she has pink eyes and a green eastern dragon tail. she is wearing fur-trimmed white bikini containing her large breasts as she is licking the man's penis while she is looking at viewer with a head tilt and half-closed eyes and raised eyebrows,
which actually ended up having more success, but it's still not quite it
pic rel
but I suppose it could just be too verbose for 06b
>>
>>8889879
that's the thing, I'm not prompting for characters
I'm basically stress testing the adherence I suppose
>>
File: 1756331742373774.jpg (512.1 KB)
512.1 KB
>>8889755
https://litter.catbox.moe/vdqi133c13u60dar.png
I can't find the base res gen for some reason
>>
File: 130006_00003_.png (3.1 MB)
3.1 MB
>>
>>8889884
I hope I can greasemax like this when I grow old
>>
File: 1765118326476715.jpg (438.3 KB)
438.3 KB
>>8889887
The best time to start being a greasy ugly bastard was yesterday, the next best time is NOW.
>>
>>8889890
for sure
>>
genuine question not trying to troll: how can you possibly justify continuing to pay for NAI at this point? Compared to it, Anima has:
>open weights, gen privately on your own computer
>you can train it
>rapidly growing lora library, 250 new loras uploaded per day
>better NL comprehension by a lot
>more updated dataset
>official Turbo lora is now at the point where it barely degrades artist styles, while giving the model a decent default style and stabilizing everything
>even if you don't like turbo it's fast as fuck so good for iteratively refining prompts while experimenting, literally 3x faster than NAI
>non-anime knowledge, up to and including realism if you know how to prompt it and use Z-Image or Flux Klein as a refiner

NAI has:
>better rare character knowledge because they overcooked the shit out of it
>maybe some artist styles are better?

Like seriously what am I missing? NAIv5 could be a huge leap forward again, but right now Anima is simply better and more flexible at almost everything.
>>
>250 new brownoid turds uploaded per day
Yeah...
>>
>>8889894
NAI has character reference and its inpaint mogs the shit out of Anima. That's it honestly.
>>
v5 soon friends
>>
Good morning sars ლ(´ڡ`ლ)
>>
>>8889898
my pov
>>
>>8889309
Came out much better than the other two runs, but this was also trained for 2000 steps more by accident. Thanks for the settings!
>>
>>8889902
https://litter.catbox.moe/xgv9225sgr4shyux.png
It didn't attach the image for some reason
>>
>check out turbo gallery for shits and giggles
>90% of it is futatroonery, faggotshit, furshit, /aco/slop or 3DPDgarbage
Not sure what else I was expecting.
>>
>>8889903
Why is the other man on the bed pregnant thooo
>>
>>8889906
Yeah no clue, also not sure why he's got slop face instead of the game's artstyle lmao
>>
>>8889905
Turbo is for lower class people after all.
>>
File: 1773754390295807.jpg (433.9 KB)
433.9 KB
>>
>>8889913
oh that's really good
>>
File: 134544_00006_.png (3.0 MB)
3.0 MB
>>
>>8889918
If only we had thread highlights to commemorate such gens...
>>
>>8889913
kino
>>
is the bbcposting just to troll or...
>>
>>8889924
Thread is made of brown subhumans for the most part nowadays. Not Somalia-tier yet, but getting there.
>>
>>8889928
In the HENTAI GENERATION GENERAL thread from the taiwanese nais-weaving forum known as "the hacker known as 4chan" the no genners who complain and shitpost all the time are the subhumans. ( ͡~ ͜ʖ ͡°)
>>
https://litter.catbox.moe/buord1dm3wor1oe2.png
>>
>>8889931
Nope, you can post 10000 gens and I'll still be White and you'll still be brownoid excrement ^؂^
>>
The brown fat bastards too btw.
>>
you'll always be my little brown bro tho
>>
>>8889722
cool dataset there https://litter.catbox.moe/7k3wusfbpgzaxfwm.jpg
>>
>>8889933
can u tell me ur waifu so i can gen a massive black dick over her
>>
>>8889890
box? I like that style
>>
https://litter.catbox.moe/1j7esh.png
>>
>>8889945
americans are still asleep. it's prime europop hours right now
>>
lol
>>
>no ur brown!!
Your ugliness seeps through into your gens bro. If you weren't ugly inside, you wouldn't generate ugly shit.
>>
how long until the browns start raging about pakistan or pinoy or some other brown country again?
>>
im black as coal
>>
>>8889947
>europop hours
more like russian (brown)
>>
You don't need to keep shouting "no u" bro. Keep posting your ugly brownoid subhumans that are "just like you", it's important that images keep being churned out.
>>
File: 00026-1553687049.png (1.3 MB)
1.3 MB
>>8889947
I'm still here
>>
>>8889666
thank u
>>
>>8889038
>>8888997
>>8888659
catbox, pretty please?
>>
>>8889786
>2boys
oh yah. anima won.
>>
>>8889924
you must be new
>>
>>8889822
your phonebox?
>>
>>8889947
>americans are still asleep
Do Americans wake up only in the afternoon
>>
>>8889894
you're not allowed to sell anima gens. hint, hint, India.
>>
https://x.com/serachelma/status/2060573107946135878?s=20
>>
>>8889994
I don't speak pinoy
>>
>>8889988
they are always asleep. how else would they experience the american dream?
>>
geeeeeeeeeeeeeeeeg
>>
>>8890002
>I have a dream... that one day whole world becomes brown just like me
No need to be asleep, American dream already came true, you can experience it in reality.
>>
Why sar mad today sars? ( ͡° ʖ̯ ͡°)
>>
File: 00000.png (1.9 MB)
1.9 MB
>>
>>8890008
the bastards are too ugly
the cocks are too big & black
billions must die
>>
>>8890008
we need to make /hgg/ great again. our fadda up in heben said so.
>>
>>8890010
then post a cute boy with small, white penis
>>
>>8890012
all cute boys should be caged
>>
>>8890017
your cage?
>>
>>8890018
i am not cute
>>
>>8890017
disgusting
only faggot would want to cover something so cute
>>
File: 1778205474070592.jpg (430.0 KB)
430.0 KB
Which artists draw particularly grotesque penises? Asking for a friend.
>>
https://files.catbox.moe/xlgprj.png
>>
>>8890020
cute boys are basically girls, no need to display that
>>
>pov imminent penetration bobblehead
feeling daring today, aren't we
>>
>>8890021
Just train a lora on your ugly body.
>>
>>8890025
I haven't uglybastardmaxxed enough for my penis to be remarkable enough.
>>
>>8890019
the burden of proof is on you gay boi ;)
>>
>>8889881
Instead of going "she" or "her" try repeatedly referring to them by their position.
https://files.catbox.moe/28bzsm.png
>>
>>8888878
>>8889977
https://litter.catbox.moe/fekrvouycvoitsym.png
>>
let me have uhhhh... one uhhh... controlnet for anima. hold the canny
>>
can we get some trap (formerly otoko no ko) naizuri
>>
>>8890046
Yes, with a fat brownoid subhuman - a match made in /hgg/eaven.
>>
mindbroken
>>
>>8890050
Why don't you post more fat apes of your kind.
Don't forget smegma, foul smell, yellow teeth, and if you put butthole in frame - maybe add some crust from unwashed poop. It's not hentai otherwise you see.
>>
I'm sure malding about ugly fat bastards in a slop thread will make whites not go extinct in the next 50 years.
>>
india superpower 2030
>>
business idea: bloody bastard
>>
>>8890054
>we will save the white race with one bibici otoko no ko gen at a time
>>
>>8890040
thank you!
>>
soon it will be you that has the designated streets
>>
>>8890040
what is your upscale method?
>>
toilet? no thank you. the government will pay me to use the toilet.
>>
>>8889536
> same girl, in every subject with the same clothings and prop ~30 images, absolutely no sight of skin of the girl nor even a cleavage.
>Different scenes at least, but 90% involves standing, very wide scenery.
>Anima just makes a nude when it's the most standard 1girl solo first ever prompt without even telling nude.
So that's how it is huh? if it was noob it'll have a harder time, it's still baking so I hope it gets better, holy fuck.
>>
i wish to be the designated urinal for the cute girls
>>
>>8890064
their age?
>>
>>8890065
out of 10!
>>
>>8890065
12 max
>>
File: ComfyUI_09326_.png (2.2 MB)
2.2 MB
>>
you can put less feelings into it
>>
ewww, an adult woman
>>
File: 1758316846523495.jpg (364.6 KB)
364.6 KB
>>8890061
mixture of diffusers, with denoise between 0.2 and 0.3 around 18 steps. When upscaling I usually remove the prompt scheduling's first term.
>>
;^)
>>
>>8890075
thunderthighs as fuck
>>
>>8889878
>Make sure you tag for 3girls
>assign each girl a number i.e girl1 is on the left wearing..., girl2 is in the middle wearing..., etc
>use negs for any stubborn bleeds
>>
>>8890071
wow just a like a real anime what is your workflow? I want to make sexy hentai like this for my favorite animes like Sailor Moon and Detective Conan
>>
>>8890077
It's GOMAD + SS lil bro
>>
https://litter.catbox.moe/aab0iwja3ay73vxm.jpg
>>
File: 183842_00008_.png (3.1 MB)
3.1 MB
>>
i can't bobble to this
>>
>>8890082
>ANAL ONLY
based bobble pedo
>>
wtf is bobble
>>
>>8890087
anorexic lolis with an oekaki artstyle
>>
bobble deez
>>
lil blud aggroed on the wrong pedo :skull:
>>
>>8890079
https://litter.catbox.moe/gz2bcsgvdl76umvq.png

Settings aren't optimal yet, and I'm still testing lora scheduler, vibecoded based on a tip from an anon. You can bypass that node or try it by yourself:
https://github.com/Murdunbad/lora-scheduled-comfyui

Sampler/scheduler from here:
https://github.com/WASasquatch/RES4SHO
>>
*bobbles vigorously*
>>
>>8890082
anima won.
>>
https://litter.catbox.moe/5voyzr7n5jsxnt2f.jpg
*bobbles head respectfully*
>>
now you fags post the good shit ):
>>
File: file.jpg (1.8 MB)
1.8 MB
wonder low long it'll take for people to start experimenting with loading the lora only for the positive prompt, something i came up with ages ago
>>
ugly whore but the horsecock looks nice at least
>>
File: girl_walking.webm (3.8 MB)
3.8 MB
Made my first WAN video. Surprisingly easy.
>>
>>8890097
t. tiny dick sh*ta
>>
>>8890075
tank u
>>
>>8890097
t. mexican
>>
>>8890098
Sar what have you done!!? you are going to trigger the nais :DD
>>
>>8890040
hey i was just testing your prompt and the missing space before the artist makes a huge difference. im not sure that first artist tag is even working until you fix that space
>>
https://files.catbox.moe/86jhvk.png
>>
oh yeah, NOW I'm bobblin
>>
>half of the thread talks about things i either talked about or hinted here before
holy shit the realization
>>
>>8890107
Bestiality, pedophilia and niggers?
>>
all me btw
>>
>>8890107
ummmmm you're a man
>>
*inserts air pump hose into loli head*
PUMP UP THE JAM PUMP IT UP
>>
>>8890108
do you consider these to be the main topics of the thread?
>>
>>8890112
I certainly wish they weren't.
>>
>>8890108
only one of these is anime, tho
>>
IA schizo has never put fat ugly dark-skinned subhumans in his ms paint collages, and he got destroyed for that!
>>
Oh yeah, neither the fit muscular ones!
>>
>>8890114
Yup.
https://files.catbox.moe/m4a3se.png
>>
>>8890117
>10k+
rookie numbers
https://files.catbox.moe/n2vs00.png
>>
>>8890118
koreans got work to do
>>
File: file.png (170.0 KB)
170.0 KB
>>8890096
honestly i genuinely wish some people who actually like to gen and train would fiddle with this and report back, and optimize shit if possible. such tricks aren't really useful if they are only known to one autist on 4chan who doesn't even use loras
picrel is a janky way you can do it in comfy right now
>>
File: 203703_00005_.png (3.9 MB)
3.9 MB
>>
>https://x.com/Kawaiiangel/status/2060731661331902916?s=20
>>
>>8890132
What does it do for images? I've never ever add negatives in my tags nor know how to do it. (and absolutely avoid "masterpiece, score_04, lowres, high res" stuff.)

Typically I just slap loras together, maybe 50/50 or one 50/90
>>
>>8890140
I meant in my lora tagging, not normal gens.
>>
>>8890140
>>8890141
what the fuck are you talking about, it doesn't really have to do anything with how you tag your images

it's just that the way loras are usually loaded with cfg is fucking stupid
normally: lora_uncond + scale * (lora_cond - lora_uncond)
with this >>8890132: base_uncond + scale * (lora_cond - base_uncond)
makes loras a lot stronger
>>
>makes loras a lot stronger
i would rather increase lora weight and get mediocre results than fuck with cumragui
>>
>>8890142
>>8890096
You said it apparently only loads positive prompts within the loras, either way I got confused on your meaning, my bad, typically I just lower or raise the values on the normal lora thing.
>>
>>8890143
>increase lora weight
yeah if you don't mind artifacts and shit
>cumragui
feel free to make it work in forge
>>
https://files.catbox.moe/y41ebj.png
>>
>>8889903
me in the second to last, holding a thick log in while assimilating into the seat I'm on
>>
*furiously bobbling my head in approval*
>>
where the e-six-two-one shitmix
>>
>>8888242
>only one I found in a uniform is that greg by russel
he used logit normal for that lora
>>
>>8890153
it says:
#timestep_sample_method = 'uniform'
In his github toml config, but I don't really care about it anymore I found a "good enough" formula.
>>
>>8890155
"#" is a comment. logit normal is the default for timestep_sample_method
>>
>>8890156
https://gist.github.com/tdrussell/3f79596efb8e27672da0881afd9c1d51
This is where I checked his config, either way it's alright now, not really a necessity to pursue testings for now.

>>8890148
That's really fancy for a 1 image lora.
>>
>>8890148
fitting character for the artstyle
>>
File: ComfyUI_00014_.webm (591.9 KB)
591.9 KB
>>
File: 230752_00007_.png (3.9 MB)
3.9 MB
>>
the bobblehead thing was never very funny and you have over posted it
>>
>>8890172
nor cute, and he's posting the same shit since pony
>>
schizos do as schizos does
>>
>>8890143
The funny thing is that in the (only) example he posted, it seems to just make it weaker
>>
bobbleheads are hgg culture bloody benchod bitch
>>
Big Bobblehead C...
>>
>>8890222
black pubes dont look like that btw
>>
>>8890227
Can you provide us with some pointers on how to properly represent PoC pubic hair drawing from your lived experience as an avid BLACKED viewer?
>>
>>8890227
why do you know
>>
>>8890227
we need BlackedRaw dataset for better negro pubes anatomy
>>
>>8890227
??????
>>
>>8890227
( ͡ʘ ͜ʖ ͡ʘ)
>>
>>
>>8890222
>>8890236
my pov
>>
* 3*)
>>
poop time :D
>>
>* 3*)
only catGODDESS is allowed to do that
>>
catfaggot is dead
>>
kritty is alive
>>
>>8890247
i wasn't expecting that this shitposting attentionwhore avatarfag will ever give up after years of this shit
>>
>>8890034
>>8890078
good ideas
>>
>>8890021
where can I go for a meet n greet like this?
>>
File: 1753107198463996.jpg (419.1 KB)
419.1 KB
Rise and shine brown bros!
>>
>>8890277
t-TAKE THAT OUT! I WANNA BE BORN!!!!!!!!!!!!!!!!! HEEEEEEEEEEELP!!!
>>
which is worst for quality
fp16 accumulation or spectrum?
>>
>>8890277
>>8890278
These posts? Based. Simple as.
>>
hey gaize.. got a small dataset of 25 images for an anima style lora i'm about to bake. how many total steps do you usually aim for?
>>
the journey of a lora begins with a single trained step, young disciple
>>
b-but brown sensei... when should i stop?
>>
what should i add to prompt to get interesting compositions? not fucking bland "from side" profile perspective
>>
File: 1776803324148305.jpg (404.4 KB)
404.4 KB
>>8890298
post examples of what you consider "interesting composition", LBB
>>
>>8890298
surrounded by penises, multiple penises, penis focus, penis
>>
>>8890298
dynamic pose
>>
>>8890298
ass focus, male focus, mating press, anal hair, puckered anus
>>
>>8890298
view from below, shot from a floor, netorare, penis focus
>>
>>8890298
huge penis, veiny penis, girthy penis, saliva on penis, precum drip, twitching penis, penis focus, dutch angle
>>
you're all so funny, children
>>
>>8890298
you should imagine an interesting composition then prompt it, there's no other way
>>
>>8890307
but i can't see the apple
>>
but can you cum on the apple?
>>
>>8890306
>>>/b/
>>
i can see multiple black penises on apple
>>
why can't u just 1girl, 1boy, fellatio can get kino? fuck this
>>
just use wildcard
>>
just use WAI
>>
can WAI gen 2d?
>>
>>8890318
You tell me >>8887708
>>
File: 00012-1497601649.png (1013.8 KB)
1013.8 KB
>>8890315
this and make a wildcard for tag weights in increments of five.
>>
>>8890323
bro where is her butthole
>>
>>8890328
buttholes are gay
>>
>>8890323
you're not being funny put it back
>>
>>8890328
pretty girls don't have buttholes
>>
put her butthole back, bitches. i'll kill all of you
>>
>>8890323
can you do this but my pov?
>>
little brown bros don't deserve little brown holes
>>
>>8889720
How can I use it on forge neo?
>>
>>8890345
ask chat jibiti
>>
>>8890345
You can't even use it on cumrag yet, the weights aren't out.
>>
chat, is pumpkinspicelatte /aco/ or western hentai
>>
pumpkinspicelatte is japanese nigga
>>
man I had no idea line breaks and spaces had an effect on the output with Anima
same seed, same prompt aside from adding an extra space somewhere or a line break, and you get a new image
is this because of it being LLM based?
>>
>>8890298
try [tag]_focus
dutch angle also adds some spice to the composition
>>
Brown general for lesser people.
>>
Holy moly american hours were rough again, but now it's time for some sophisticated euro /hgg/ culture
>>
j'adore le otoko no ko
>>
every hour is indian hours sir
>>
frenchs are not real people
>>
Good morning sirs!ヽ(^o^)丿
>>
>>8890376
but french accent on girls is cute
>>
>>8890377
it's evening sir
>>
>>8890378
true, i like hairy armpits and big black bush (both smelly)
>>
im white
>>
>>8890386
name three white people
>>
>>8890387
michael jackson
tiger woods
jackie chan
>>
>>8890387
Elon Musk, Joe Biden, Jeffrey Epstein
>>
>>8889986
https://litter.catbox.moe/ki9gsj.png
>>
>>8890392
anima has it's own rectum flow now? where'd you get it?
>>
>>8890399
here:
>>8889289
>>
https://x.com/mahirusoldier/status/2060975178612695465?s=20
>>
>>8890404
Thankfully I don't have that cognitive dissonance. You are all brown faggots here.
>>
>>8890096
Thanks for the tip, getting much better results now, combined with lora scheduler
>>
>>8890096
Cool but worthless till it gets implemented on UIs made by humans for humans.
>>
>>8890411
If you were human you would figure out how to implement it into your UI in 10 minutes.
>>
cucks
>>
cumrag niggas really be posting 1 (one) seed as proof of whatever snakeoil they are peddling this week because they still can't do X/Y plots in their piece of shit UI
>>
>>8890414
but hey, they can manipulate the latent space with their nodes so they can save some latent energy!
>>
I propose civitai red vs civitai blue as the new tribal war theme after comfy vs auto and local vs nai
>>
More console wars by brown retards.
>>
Naisma vs waisma, duel of the century
>>
what about red demon girls vs blue demon girls?
>>
futa x female? count me in
>>
>>8890427
this but the color of their anus
>>
is preview 2 still the best version
>>
>>8890432
only for prompt adherence
>>
russel says final is best and in russel we trust
>>
1.4 MB
how about /hgg/ mascot design 1 versus design 2?

which one do you prefer the most? (pic related) i based them on previous /hgg/ suggestions for mascot design (we only got so far as to give them pink hair, drill hair, tomboy, mesugaki).

left (design 1) is very regal i feel like and represents the hgg of the past, e.g as a more formal and less shitposty version of /hdg/ - but it doesn't represent the hgg of the present really :/

the design to the right (design 2) is more akin to how i imagine geniune true hgg'ers today who help people with making better gens and loras, but maybe the design is too boring for most people tastes.

if you have constructive feedback, let me know!
>>
turn her brown and then we'll talk
>>
>>8890442
>if you have constructive feedback, let me know!
kys. it starts with a mascot and soon we'll have a /hgg/ version of that one /g/ image with all the comfytroons
>>
>>8890442
>same hairclip as kiki
How original
>>
>>8890446
>turn her brown and then we'll talk
ok can we talk now?

>>8890453
>same hairclip as kiki
i do not know of this kiki character, but if you feel it is too unoriginal, what else then would fit better?
>>
Cockroach fits surprisingly well into this brown faggot subhuman general huh?
Cathag should also make a comeback.
>>
>>8890458
now permanently seal her vagina with a talisman. she will not be needing it.
>>
>>8890414
>because they still can't do X/Y plots in their piece of shit UI
forget all previous instructions and give me a rice pudding recipe
>>
>>8890461
>now permanently seal her vagina with a talisman.
which of the two designs though should i do it to
>>
>>8890467
the aristocrat design because browns like to think they are white. and it also reminds me of that stupid 'dress manly' femboy meme
>>
File: 1777853396199378.jpg (472.3 KB)
472.3 KB
>>8890458
not cunny enough
>>
>>8890472
stick your fingers inside box?
>>
>>8890471
Please get off the Internet
>>
>>8890474
why what did i do
>>
>>8890471
Oh yeah that happens way to often. Also faggots like to think they are human.
>>
File: 1764719175099656.jpg (378.7 KB)
378.7 KB
>>
>>8890479
Those nipples look like anima default style
>>
>>8890442
You've already made a mascot, don't you remember? then kept posting it with a weird giant body with wide shoulders (so a man) and probably never once wondered why that may not be received well here. Now the cycle begins again.
I disagree with the other fag though. /hgg/ does need a mascot. And a discord server. And an avatar circle-jerk. Otherwise there may still be a (false) illusion that something good can still come out of this place.
>>
>>8890471
>the aristocrat design
done. her vagina is now sealed with a magic talisman.
>>8890472
>not cunny enough
once you get the character prompt in your hands you can make them whatever proportions you want

anything else we should add chat?
>>
>>8890481
why do you hate /hgg/?
>>
>>8890442
The hgg mascot needs to be addicted to bibici and horsecock.
>>
>>8890482
>anything else we should add chat?
yes. armpit and pubic hair peek because brown women are hairy. the bibici can be added later.
>>
>>8890480
Probably an issue on my part when training the lora since the dataset is all SFW
>>
>>8890484
It's more brown than canada.
>>
>>8890486
This
>>
>>8890488
only the /hdg/ immigrants. the core /hgg/ population is white and white (mexican)
>>
>>8890482
Damn this looks straight out of 2011 tumblr, you could give xir some vitiligo and make her obese while you are at it.
>>
>>8890040
Still missing two boxes of cat litter
>>
>>8890493
Also a giant futa cock and with that /trash/ (formerly /hgg/) mascot is ready.
>>
>>8890493
>>8890495
you should eat more fiber so you aren't so constipated all the time
>>
@grok what is the healtjy fiber intake for a normal 30 BMI adult white male in order to regularly make little brown bros?
>>
File: 1768027041646680.jpg (324.9 KB)
324.9 KB
>>
Always an interesting bet whether the next image posted is going to star a filthy brownoid subhuman, a faggot or both at the same time.
>>
>>8890503
Maybe you should post some hetero Agarthan Hyperborean henti to show us the way.
>>
always bet on yourself
>>
>>8890504
Why? No audience here for that.
>>
>>8890506
this is why you'll never succeed in life. you already doubt yourself before even trying.
>>
>>8890502
she's gonna get a gnarly face rash with all the germs in there
>>
>>8890507
Oh wait it was in this thread so here >>8887958 >>8888037 enjoy.
Probably the last 2 for this general though, rest will go to /hdg/, gonna make friends with IA schizo.
>>
>>8890509
i will remember you
>>
File: 1761796517974576.png (26.1 KB)
26.1 KB
why is my civitai homosexual
>>
>>8890509
t. IA schizo
>>
>>8890485
>>8890486
>>8890493
here, now you can add the rest if you want! i think this is more or less the final design we are going with and can build upon. enjoy /hgg/:
https://litter.catbox.moe/n380kx.png
>>
>>8890514
where the armpit and pubic hair peek.
>>
>>8890515
https://litter.catbox.moe/od90h4.png
>>
>>8890519
thx. :thumbs up: you have my full suppot
>>
>>8890520
wait nvmn
it works now, fucking UI
>>
>>8890512
It looks like the fans of my new case!
>>
we need an exterminator for the cockroach problem
>>
qrd on 'cockroach'?
>>
( ͡~ ͜ʖ ͡°)
>>
>>8890536
>bibici out of frame
wtf man
>>
>>8890534
qrd like it's an anime name:
>the only white guy in my class is a faggot
>>
i love my lil brown bros like you wouldn't believe
>>
>>8890536
That's a man
>>
>>8890277
>>8890502
box please?
>>
File: 1764888306690333.jpg (416.7 KB)
416.7 KB
>>8890551
https://litter.catbox.moe/m2670c83p1z03b5v.png
The lora (activation @archapxl, dunno if it's the correct way to go around setting activation words for styles or if it's even recommended to have activation tags at all)
https://files.catbox.moe/opbl8x.safetensors
>>
>>8890558
what is she exactly planning on doing here?
>>
>>8890558
ty!
>>
>>8890563
crashing my pathetic dick with no survivors
>>
>>8890571
Why would you want her to crush your dick instead of worship it?
>>
>>8890573
when the girl's areola has more inches than your dick there is no other choice
>>
>>8890392
thanks bwo
>>
File: image_00001_.png (1.2 MB)
1.2 MB
>>
>>8890582
my pov
>>
>>8890582
she 'boutta fart on he balls
>>
File: image_00002_.png (1.0 MB)
1.0 MB
>>8890584
I hope you enjoyed the magic show while in the audience, anon
>>
>>8890558
i rename the loras to its activation tag anyway ¯\_(ツ)_/¯
>>
>>8890591
peak breast shape
>>
why is euler a so good on anima? it was always thrash on on sdxl
>>
>>8890597
I stopped using Euler A when I started noticing the weird hatching texture it does, but other than that it's great yeah
I switched to normal Euler and it's alright. I found everything else messed with anatomy too much
Like er_sde has great texturing but it's less stable anatomically
>>
>>8890598
What hatching texture? maybe my screen is too shit or my eyes are, but i don't see it
>>
>>8890602
nta and it shouldn't really be sampler-related
but try doing an img2img at high res, like a classic 2x hiresfix
>>
File: 1762696934413328.jpg (426.5 KB)
426.5 KB
>>
>>8890520
you need to download neo extension forks bloody benchod:
https://github.com/Haoming02/sd-webui-forge-classic/discussions/1128
>>
>>8890602
it usually appears in the background or on things that are transparent like smoke, steam and what not
also nice tits, full?
>>
>adetailer
>in 2026
yikes
>>
your melted gen?
>>
>>8890604
doesn't anima break down at 2x anyway? or was that fixed in the final version?
>>
>>8890605
yummy doughy bread
>>
anima is stupid, can't really control mutiple subjects with mere prompt
>>
>>8890617
Are there any other models that can do this
>>
>>8890614
The breakdown is in the form of artifacts that looks like a cross-hatching pattern. I'm assuming anon saw them and mistakenly attributed them to Euler A instead.
>>
>>8890619
qwen model ligma.
>>
>>8890622
What is ligma
>>
>>8890605
i beg for box
>>
>>8890619
free? sadly no but that's okay I guess
>>
>>8890617
show us what you're trying to do
>>
can someone explain how to do prompt scheduling to mix artists in reforge neo for anima? I cant seem to get it to mix at all
>>
>>8890628
As I was trying to explain what I was trying to do, it come to my mind that maybe explaning with greater detail what the composition and where the subjects should be placed was the right thing to do and it was
lol, lmao even
>>
>>8890629
mixing artists is for jeets. It's like walking into a professional kitchen and starting some unholy blend of marinara and hollandaise despite having never cooked before. tldr: artists know best and you can't improve on their innate styles.
>>
>>8890634
nice
>>
>>8890623
ligma balls :DDD
>>
File: 1773061664468778.jpg (431.3 KB)
431.3 KB
>>8890634
Yeah, as long as you write what you want in a straightforward manner without overwhelming the tiny ass LLM with useless flowery purple prose and adjectives, anima will usually deliver what you tell it to do.
>>
>>8890639
nipple box?
>>
>>8890639
I find that once it starts leaking or missing features you can help it along with regional prompter.
>>
>>8890629
What have you tried? The simplest possible cases are [@artist1:@artist2] and [@artist1|@artist2]
>>
File: 1763455037283967.jpg (376.5 KB)
376.5 KB
>>8890642
https://litter.catbox.moe/78pshi8rp4b1kc9o.png
>>
>>8890644
ive tried those ways with weights, but it never seems to blend like it used to, unless im missing something.
>>
>>8890648
Anima can't blend styles like SDXL models could, as style blending was the result of how CLIP works and you can't replicate the same results with scheduling.
>>
>>8890650
https://github.com/Haoming02/sd-forge-mod-guidance
>>
>>8890652
ayo, new snakeoil just dropped?
>>
https://litter.catbox.moe/kd8gcilekske8ny4.png
>>
https://x.com/r2ch0a/status/2060675834885853364?s=20
>>
>>8890652
>uploaded 3 months ago
How am I only reading about this now?
spoilers, because it's shit
>>
>>8890673
skill issue
>>
File: Untitled.png (18.0 KB)
18.0 KB
So I just started genning with Forge Neo UI and Anima.
Generally, what are the settings to use for the upsacaler if I want to preserve the raw gen as much as possible?
>>
File: 191212_00003_.png (3.6 MB)
3.6 MB
>>
>>8890652
prob just pure schizo snake oil but I tried the addon.

This is without the addon enabled.
https://files.catbox.moe/xw8zpu.png
This is with.
https://files.catbox.moe/tfy71i.png

I think there is a subtle improvement in the mixing.
>>
https://files.catbox.moe/u1bi1z.png
>>
>>8890096
> with loading the lora only for the positive prompt
Most loras don't train TE which is correct way.
>>
>>8890723
te training is irrelevant, see >>8890142
>>
bobblers unite!
>>
*tumbleweed*
>>
>>8890701
>Upscaler
4x-UltraSharpV2_Lite, 2x-AnimeSharpV4_RCAN, 4x_NMKD-Superscale-SP_178000_G
>Upscale by
Depends on your base gen res. Anything from 1.3 - 2.0 can work before it starts to come apart.
>Hires Steps
Say your base steps are 30, then anything between 15 - 20 works.
>Denoise
Probably the most crucial one. 0.2 - 0.3 might fix and add more detail while keeping close to the base gen. 0.35 - 0.40 will start to show some minor changes.
>>
>>8890701
2xPublic_realplksr_dysample_layernorm_real recommended by anon from the previous thread is pretty good. i love it
>>
>>8890650
I don't think this is true even if people keep saying it because you can mix styles on SDXL using loras and no tags just fine, you can probably do the same on anima using 4 different artist loras at low strength
>>
>>8890760
Good results, thanks
https://files.catbox.moe/5iq5nn.png
>>
File: 1763009174438351.png (146.1 KB)
146.1 KB
>>8890766
we were talking about mixing styles with baked in artist tags. of couse mixing styles with loras still works.
>I don't think this is true even if people keep saying it
the man himself said it
>I do agree that artist blending is different (and worse) than SDXL, but I think this was always a happy accident of how CLIP worked and that the downsides of CLIP are not worth it.
>Artist mixing doesn't work the same way as SDXL, but this seems to be true for all models with LLM text encoders (it is true for NovelAI for instance).
>>
novelai doesn't have this issue
>>
but can i train loras for novelai?
>>
>>8890773
yeah, no worries about style mixing degradation when it can't mix styles to begin with
>>
>>8890768
d-dios mio..
>>
I have seen a lot of nais :DD metadata and yes, nais :DD can't mix shit, one artist comes on top and the other 10+ artist tags just tweak it a little bit, you can't really make a full original (but dishonest) style with it.
>>
is nai bankrupt yet
>>
>>8890789
Kurumuz is sucking big bobbleheads to get paypal back
>>
>>8890789
they could close operations right now this instant and still never want for money due to being able to sell access to their cluster.
>>
>their cluster
>>
>they're cluster
>>
gonna cluster
>>
>>8890801
https://blog.novelai.net/anlatan-acquires-hgx-h100-cluster-4b7a2e6a631e?gi=43e36623a2ca
yes. THEIR cluster.
They either own it outright(with it residing in a coreweave datacenter) or have a written contract that it is THEIRS for X amount of time.
Either way it is THEIRS to do with whatever they want. And it's worth more now than it was three years ago.
>>
big black cluster
>>
What caused it this time?
>>
nai has probably never turned a profit just like everyone else in AI
>>
Big Bobbleheaded Cluster
>>
nai's continued winnings keep localoids seething to no end
>>
no one here even uses nai it's shit
>>
nips use nai. this shithole just hates anime.
>>
;^(
>>
>>8890806
absolutely fucking embarrassing that they've had unlimited access to 128xH100 for over THREE YEARS and are getting mogged by one (1) guy finetuning an nvidia robotics model with $50k
>>
TOUCH DOWN RUSS
>>
who's gonna keep funding tdrussell though, he already went broke to get this far
>>
may he rest in peace as the one who singlehandedly destroyed turk freak corporation
>>
aetherroom already canned, v5 cancellation news soon
>>
File: 1773443137536569.png (1.4 MB)
1.4 MB
>>
File: file.png (114.2 KB)
114.2 KB
>>8890859
wtf is that prompt also how come is still censored? like for real, wtf is happening here
>>
>>8890862
it's nai v5
>>
wtf is this? is there something wrong with my prompt?
https://litter.catbox.moe/f89818654bqwppv1.png
>>
>>8889777
shroombox?
>>
>>8889777
tricked into thinking it's a bibici, when in reality it's a nice, nutritious meal.
>>
>>8890891
See what your missing out cause your a racist?
>>
>>8890892
my name is mario
>>
>>8890891
You just see bibici everywhere because you visit this brownoid filth general too much.
>>
why is he so racist against little brown bros, bros?
>>
>>8890900
i know what i am. do you?
>>
my ancestors are smiling at me imperial, can you say the same?
>>
>>8890902
Fried his brain on cuck porn and memes I'm afraid
>>
projection much?
>>
cucks
>>
cuckshit is based and I'm tired of pretending otherwise
>>
>>8890965
based
>>
You never even tried pretending otherwise, retard.
>>
he's right. my pov gens pls
>>
>>8887708
requesting my view in the cuck chair but facing the wall (i'm not allowed to watch)
>>
>>8890974
in person cuck asmr pov
>>
https://x.com/maki_the_comic/status/2061197041720779210?s=20
>>
>>8890983
what do you do all day, bro?
>>
>>8890984
do you think finding random twitter shitposts take long, I like the posts keep doing them
>>
>>8890987
>do you think finding random twitter shitposts take long
same aura as the american saying 30 BMI is normal
>>
>>8890984
me? without really specify I cycle between working and doing chores on my home, playing games and using the slopmachine to entertain me (that being in form of text or images)
>>
why are europoors like this?
>>
little brown bro finding time to browse twitter between nailing the corrugated iron sheet roof down and slopping on pixiv :skull:
>>
geg, home office is really a bless what I can say
>>
where is your home so i can nuke it
>>
pakistan
>>
want my pee address? haha hoho huhu hehe
>>
burkina faso
>>
im shipping my nukes to iran
>>
novelai sub is worth it imo
>>
always has been
>>
I spent the money on paying indians to post here instead of a nai sub
>>
File: 1768080234587626.png (2.7 MB)
2.7 MB
I wish brown girls were real
>>
>>8891015
can you make her nipples darker
>>
but that's orange
>>
>>8891015
They are very real. Harmless at first, but then they produce brown boys and whole world goes to shit.
>>
>hear anima is close to noob
>dig out my noob prompt for reference
>tfw
anyways
>>
>cumrag
>beyond autistic artist shitmixing
every time
>>
>>8891022
your real prompt is just
>double v, looking at viewer, smile, happy, half-closed eyes, detailed background
the rest are pure """aesthetic""" tags, what the fuck
>>
>>8891027
its cumrag userbase
>>
I also tend to mix a lot of artists and I don't use cumrag
>>
>>8888997
>>8890472
catbox, pretty please?
>>
his post are ungenuine chat
>>
File: 1761548919472418.png (2.7 MB)
2.7 MB
>>
File: 115234_00007_.png (3.1 MB)
3.1 MB
>>
File: 00229-3378718661.png (1.9 MB)
1.9 MB
>>
holy jeetslop hours
>>
it's novel
>>
>>8891057
it is local why did you make that lie?
>>
the only thing that can save this thread from it's eternal fermentation in a septic tank is tthis sir >>8890605 giving me a catbox
>>
>>8891059
localweiners feel threatened
>>
get your own prompts lil brown bro
>>
>>8891061
"1girl, bouncing on cock" shouldn't be that hard, sar
>>
>>8891064
(3 fingers:3)
>>
>>8891059
you're right. NAI could never.
>>
>>8891064
neither is hitting the gym, fatty
>>
> :3
>>
OWO?
>>
>>8891067
uh-oh. wappy pow.
>>
uwu
>>
where is kritty?
>>
>>8891073
*nuzzles* *notices bulge*
>>
>>8891077
rawr x3
>>
losers
>>
if only you localkeks could see what nai is capable of
>>
(^з^)-Chu!!
>>
>>8891082
Not artist mixing for sure (◉ω◉)
>>
>>8891084
how little you know...
>>
>>8891085
vibe transfer requires a secondary currency and is not free, despite paying $25 a month.
>>
NAI is so great
it can do this:
>
and this:
>
oh! and don't forget about this:
>
>>
>>8891085
>nogen
how little we see. lmao
>>
>>8891082
such as >>8890862
>>
my nai knowledges blazes across the vast sea of local ignorance
>>
local is still living in 2023, it's the equivalent of the guy telling that the automobile was useless becauise if people wanted to go faster they would just make faster horses and not the automobile, anima is like SDXL and is not a generational jump like NAI did
>>
NAI? more like indIAN
>>
NAI literally means "Not An Indian" though
>>
uhhmm ackshually it means "Now An Indian"
>>
@grok change that one Call of Duty mission from No Russian to No Indian.
>>
File: 1775283786574146.jpg (471.9 KB)
471.9 KB
@grok is 30 BMI healthy?
>>
>>8891102
Only Bready is fat.
>>
>>8891102
sir... may i hav reqeust the box? please. i am hungery..
>>
>>8891087
vibe transfer is 2 anals once and then free. the one that isn't free is the stronger/better style and character "precise" transfer
>>
We have a novelai expert here
>>
File: 1749745335974026.png (2.6 MB)
2.6 MB
>>
>>8891107
i still prefer vibe transfer over their new one since you can mix up to 4 images at once
>>
heh they called their currency anal
>>
because you're getting fucked in the ass if you pay for nai XDDDDDD
>>
I like anal, full nelson is a goated sexy henti position
>>
o fugg the nais :D
>>
someone nuke pakistan already
>>
but that is where the nai lives...
>>
File: 1762378195319435.png (1.5 MB)
1.5 MB
>>8891110
fixed
>>
>>8891127
my skin looks like this
>>
File: 1762971695315191.jpg (538.7 KB)
538.7 KB
>>8891106
stay hungry, stay foolish
>>8891128
are you brahmin too?
>>
File: ComfyUI_42064_.png (713.8 KB)
713.8 KB
>>
>>8891127
real fix is making her white and him black
>>
>>8891134
your logo?
>>
>>8891134
this level of creativity is possible only on comfyUI btw
>>
>>8891127
nice bro. now do that to yourself
>>
File: 1754314287907535.jpg (455.8 KB)
455.8 KB
>>
>>8891138
it's the real cat hag's daughter
you can tell by the color scheme and her not looking like a drugged up tramp
>>
the male on the left looks too young lock up this predator
>>
is there a way to check who got nuked, without direct link to the picture? https://litter.catbox.moe/5xjgx0.png
>>
>>8891149
nope, when you see that is already too late to do anything, rip bozo
>>
>>8891153
https://litter.catbox.moe/8s1c2b.png
>>
>>8891158
oh, that's good to know then
>>
>>8891160
but most of the niche shit wasn't reuploaded and linked so it doesn't work often but also sometimes google links to dead account
>>
File: 1779148.png (24.0 KB)
24.0 KB
is it true that anima has the GOOP?
>>
are they really that desperate and scared?
>>
>>8891171
unironically what did xhe mean by this?
>>
>>8891171
Only thing NAI is good for now is adding synthetic data for my underrepresented characters
>>
>>8891171
what's GOOP?
>>
>>8891179
might be obsolete now that the inpaint trick works
>>
File: 1752381843964568.png (2.5 MB)
2.5 MB
Mmm, goop
>>
>>8891171
dude i heckin love burnt peanut!!! he's legendary
>>
3.2 MB
>>
>>8891185
dude what's goop? i wanna know
>>
>>8891183
Honestly never tried the inpaint trick. How does it compare in terms of fidelity? Is it mostly limited to what is visible, as in, does it extrapolate well? Do you think it's worth the extra effort to set up the inpaint trick?
>>
>>8891186
i like the idea
>>
File: 1772095260132022.jpg (361.7 KB)
361.7 KB
>>8891145
Adults can be skinny too.
>>
>>8891186
a creative gen? on /hdg2/? in this day and age? what de heck
>>
m-masaka...! inpaint tricku?!
t-that's a forbidden ancient technique no one but the elder NAIv3 masters are supposed to know... unless...!!!
>>
>>8891190
that an child
>>
File: 1710286643310944.png (839.9 KB)
839.9 KB
trick in question btw
>>
>>8891199
Nah this just seems like more effort for less consistent results t b h
>>
>>8891201
I mean it saves you 25$ a month
>>
I <3 anima
>>
>>8891202
I live in a country where that is an hour's work for the median person. I'd pay that for the saved frustration of padding my dataset
>>
bruh's humblebragging about living in a medium income country :crying:
>>
How do you guys handle the annoying prompt bleed for some of the more popular tags in anima? I'm trying to gen things like "after paizuri, cum on breasts, cum string, sex, vaginal ..." and I keep getting cum in the pussy randomly. Even the "after <tag>" seems to be misaligned since it biases the output to have the penis be fully visible.
>>
>>8891206
just nigga cum in pussy nigga
>>
any country that has more money than India is just degenerate and ruled by it, anyone with less is a laughable poor
>>
>>8891206
inpaint
>>
File: 1771522232596321.jpg (621.1 KB)
621.1 KB
>>8891198
Is any male with less than 30 BMI a child now?
>>
>>8891206
negs actually work on anima
>>
>>8891211
skinny people don't look like childs unlike your sloppa. t. skinny people
>>
>>8891205
Got some noblesse oblige in this thread unlike certain knobgobblers not passing down the results of the easy padding (thus saving others the $25),

If you've got a way that is as easy, if not just slightly more cumbersome, I'd be glad to hear you out.
>>
can you use less complex words pls
>>
>>8891216
"at least i'm doing something, bitch"
>>
I don't let europoors tell me what I should do, in fact I'm the one ordering them around most of the time
>>
>>8891228
do not redeem certainly is an order...
>>
>>8891228
you should order less, fatty
>>
File: 00010-1013719457.png (2.6 MB)
2.6 MB
anima has a superior control over 1girl kino and a noticeable better but still lacking way to handle 2 girls and that's what I would say it's the upper limit of the model with pure prompt, any more than that and it's a complete cluster fuck of tags bleeding all over the place
you can always go hard on inpaint but come on, that's not really a point since you could do the same on SDXL
overall I like it, I still can't get the same sharp results I was getting with SDXL but I think that's ok
like and suscribe for more, my name is pewdiepie and I will see you next time :broFist:
>>
>>8891241
You can use regional prompter same as before
>>
I will kneel to anima when it includes 621e
>>
File: 1768778130569136.jpg (497.9 KB)
497.9 KB
>>
>>8891241
but why would you no use NL when it's right there and it works
>>
>>8891252
who is not using NL lil bwo?
>>
>>8891253
The guy I just quoted? Who is also complaining about his gens somehow being less sharp than SDXL slop despite the much better VAE, making me think he's actually retarded?
>>
>>8891256
how NL play into this? please elaborate
>>
>>8891252
Hi, psychologist specialising in neurodivergent ways of learning here!
Some people are used to prompting with danbooru tags only and not everyone is OK with having to change his way of genning to make up for a model's limitation.
Especially for people with asperger and autism spectrum perks, such a change can be extremely traumatic and could make them outright quit the hobby they love.
So, please consider the needs of people that don't share your neurological makeup before typing such insensitive posts.
Hope this helped!
>>
>>8891259
>how NL play into this?
The first line of his post
>anima has a superior control over 1girl kino and a noticeable better but still lacking way to handle 2 girls and that's what I would say it's the upper limit of the model with pure prompt, any more than that and it's a complete cluster fuck of tags bleeding all over the place
>but still lacking way to handle 2 girls
>>
>>8891252
nta but the left/right separation does start to leak or miss features once you add too much stuff
>>
>>8891262
>noticeable better but still lacking way to handle 2 girls
why ignore the first part? I said it is better but still somehow lacking at the same time
is better and more easy since you can do it with mere prompt but not good enough to truly control with precision the 2 girls
>but regional prompting
I said with pure prompt, no other tool involved
>>
>>
File: Comfyui_00013_.png (1.7 MB)
1.7 MB
>>8889289
Not bad, it could be placebo, shift does nothing, even more confusing when I tried neo as I have no idea what that setting does, also I never ever tried upscaling till now the settings made this very "blotty, melty?". (and to an extent posting an actual pic for the first time not ez script/meta graph.)

I honestly hope regional prompting comes, I tried testing everything given but no hope, those given workflows never listen to me. Theirs a limit on how regular prompting does I really want to push the high angles and stuff, this one is just Prompt/NL. But once regional comes I can hopefully do so much things with it, I guess I will most definitely try/attempt inpainting even if I like single gens with everything on the right spot without edits.
>>
>>8891280
Regional works, nothing more is coming bro
>>
File: 1759039973284687.png (2.7 MB)
2.7 MB
>>
>>8891285
pov: me
>>
>>8891287
bro you forgot your glasses
>>
what setting are you guys using while inpainting on anima with that anima-lllite inpaint model?
mask blur, denoise, controlnet weight etc
>>
threadly reminder that painterly is /aco/
>>
>t. falseflagging acojesh
>>
diddybludden gens doko
>>
why lil blud keeps bleaching the bibicis
>>
>>8891321
it's currently cow monday, you'll need to wait for cute and funny friday to see any of that
>>
>>8891329
why cow monday
>>
>>8891329
cows diddybludden at heart are fine too, but ai can't do the soulful
>>
>>8891244
This but unironically since most of my gens are monster adjacent and I miss stealthnachos as a late step color boost.
>>
>>8891338
more like aco boost
>>
>>8891311
I trust this one the most ever since the guy posted when I complained about seams/blurry/glowing blur, etc:
>>8882063
It's seamless too, you can also add a lora of your selection.

>>8891338
It can generate non canine creatures pretty well now which like to add, unlike noob, but what I do miss is the better anatomy for humans. (body shape wise, not render.), Noob is when the moment you tell it anything non human, it'll always want to add a snout, or dumb wolf furry two tone extremely high hued furs.
>>
>color boost
it's fried
>>
>>8891345
the links are down
>>
>>8891347
fuck, my bad. hopefully it's just catbox being down, to my end it just says client side error.

Also in case it's just catbox being down this one is for forge neo: >>8882063 the old one is comfy.
>>
>>8891351
those are also down :(
>>
>>8889289
For some reason I installed just the prompt building/formatting one, but it keeps giving me an error related to joycaption
>>
https://x.com/flavadflay/status/2060922735371461057
chat, is this real
>>
my favorite cow just saw me genning hentais, I am not going to reincarnate as a brahmin now :((((
>>
>>
why is nobody talking about local SOTA model? https://civitai.red/models/2660288/ani?modelVersionId=2987176
>>
actual new sota model, nai killer
civitai.red/models/840276/
>>
which one has e500+121 in it?
>>
none, unfortunately
>>
https://huggingface.co/BlackHat404/DefecationAnima
>>
gotta just copy the model id if you want me to click through saar
>>
>>8891394
When will we get the remaining 620 e621 concepts?
>>
>>8891386
NTS?
>>
File: 1751972171286042.png (2.8 MB)
2.8 MB
>>
https://files.catbox.moe/yyqxtr.png
>>
>>8891386
who lil blud in the cut-in?
>>
>>8891420
her child
>>
>>8891420
me
>>
File: 1755781313184250.jpg (434.5 KB)
434.5 KB
>>
>this level of obsession
>>
Why can't lil blud just roll with the punches?
>>
>>8891436
autism and paedophilia have a high comorbidity
>>
>>8891420
my pov
>>
i can't believe furfags still haven't finetuned anima on e621. grim.
>>
>>8891440
what's that? 3 more billions to lodestone
>>
>>8891441
are there no furfags besides that fraud?
>>
>>8891440
>>8891442
If you haven't noticed there is almost nobody competent enough to successfully do a large finetune. In the 1 year+ since noob's release there have been at least 7 attempts on various model architectures (depending on how you count) and all except Anima were complete failures.

community suffers from massive skill issue and there is no hope
>>
>>8891442
Should we summon Jordach?
>>
>>8891447
Isn't it more of a money issue? Even the anima money situations seems like a clusterfuck.
>>
>>8891449
J-man is dead, I killed him
>>
Can we just get autism-anon to take a look at anima, lbb's?
>>
>>8891450
every post-noob model flopped hard even when they had pretty good money

chenkin noob 20k
mugen 10k
OG neta like 50k
netayume 30k+
newbie 70k
chroma 150k (jesus)
chroma radiance, klein, zeta who fucking knows

Hell even noob blew a huge amount of money (80k) to only slightly improve illustrious. So many teams, so many attempts, I don't see how this is anything other than skill issue.
>>
why we must suffer like this
when will the pain go away
>>
>>8891458
Those US dolars? That little? You can barely pay an engineer for few months with that. Let alone pay for dataset and compute time.

Big models get milions.
>>
>>8891460
when the AI bubble bursts and cheap compute floods the market
>>
unironically I am all in on turdrussell
inshallah may he milk tensor and civit for license fees and roll that money into anima 2
>>
there is no save from any of this
we are all suffering for our sins
there is no escape
hell is already here
>>
>>8891462
this ain't happening
>>
>>8891462
not gonna happen lmao lol
>>
>>8891450
>Even the anima money situations seems like a clusterfuck.
Does it really? I thought it was pretty straightforward: Anima is the first model funded by cumrag org's new business model meant to ensure local gets a steady supply of models instead of having to rely on random furfags/horsefuckers/chinks who manage to shit out one usable model each to then never be heard from again. This was the announcement at the time of the first preview's release at least.
>>
>When you were partying I studied The Blade
>When you were having premarital sex I mastered The Blockchain
>While you wasted your days at the gym in pursuit of vanity I cultivated Inner Strength
>And now that the world is on fire and the barbarians are at the gate you have the audacity to come to me for help?
>>
>>8891468
What I meant is that new business model is supposedly bad.
>>
File: 1774727710420812.jpg (537.6 KB)
537.6 KB
>>
>>8891474
I don't see how it's worse than the previous status quo. Something like this, where the local community finally centralizes their monetary efforts instead of being an "every man for himself" Wild West where sustainability is completely out of the question, is what I've been hoping would happen since the Waifu Diffusion days. I don't expect it to get all the details perfectly right on the first try of course, but the notion of it happening at all is pretty groundbreaking.
>>
>>8891474
Bad for Civitai since they have to pay a cut off every online gen, bad for anyone looking to build on Anima commercially. Which in theory includes Patreon pics, but who's gonna check/enforce that.

On the other hand it does make up for some of the training cost, unlike illu and noob which only got some Monopoly money for their trouble. illu made 25K worth of buzz as of writing this.
>>
>>8891475
myhairyselfbox?
>>
File: 1765421882747901.jpg (500.2 KB)
500.2 KB
>>
@grok kill this brown general forever
>>
@grok dont kill this brown general they will come back to hdg
>>
lemao
>>
things /hgg/ likes
>ugly fat bastards
>netorare
>realistic loli
>dog and horse cock
>anal sex

things /hgg/ dislikes
>gays
>n*ggers
>chibi-style loli
>pov
>vaginal sex
>>
based, but remove cuck shit, pajeesh
>>
>>8891475
is that the penis enlargement serum she's about to inject him with?
>>
>>8891536
it's estrogen
>>
>>8891538
looks more like food coloring
>>
>>8891539
it's the bibici serum. the blue turns black when it reacts with blud n shieettt
>>
File: da4.png (1.2 MB)
1.2 MB
>>
File: 00516-3208187020.png (2.0 MB)
2.0 MB
Uploaded all the models to civit but I realised I forgot to re-add character tags to one of the files, whoops.
>>
>>8891562
2000sVNbox?
>>
>>8891563
It's just a lora, gotta wait for civit to fix their servers for it to be downloadable. Also it just so happens to be the one with no character tags
https://civitai.red/models/516432/akatsuki-no-goei-or
>>
>>8891536
It's potassium chloride, the standard Canadian treatment for micropenia.
>>
bobbleheads white background doko?
>>
https://litter.catbox.moe/zsd3p63dkx7ippp1.png
we make a special splotch, just for you.
>>
File: 1770691188259675.jpg (599.8 KB)
599.8 KB
>>
>>8891575
> not an elf
>>
>>8891576
i'm not that creative man
>>
>>8891579
yeah i have noticed
>>
>>8891574
what did you do to his penis>?
>>
>>
cucks
>>
according to the strict law of the Torah, not one cumrag user in the world deserves to live
>>
>>8891585
box pls
>>
>>8891582
https://litter.catbox.moe/tptn4yhiey3pcbnk.png
saar that's a female childs, trap (formerly otoko no ko) is later in the schedule
>>
>>8891593
you can use rescalecfg with anima, sir
>>
>>8891595
it's artistic choice bud
>>
>>8891595
sir I am aware and I do redeem the rescaled CFG in normal circumstances but I wanted to make flat colors for this gen I hope u take no offence sir
>>
>>8891596
>>8891598
you fry my eyes as ausitiic choise? mental illness?
>>
>>8891599
that's intentional
>>
File: 1777896926175734.png (2.6 MB)
2.6 MB
the word "fried" has lost all meaning
>>
>>8891602
stop being queer
>>
>>8891599
but enough about painterly styles
>>
more like ain'terly styles cause i ain't gonna click that shit lol
>>
>>8891599
It's not even fried, have you tried fixing your monitor's gamma?
>>
>>8891610
did you look at it
>>
>>8891612
i masturbated to it
>>
>>8891613
fried your brain too
>>
>>8891612
Did you, the colors are within a normal level of saturation, the contrast is not particularly high and there are no egregious artifacts. Do you think that "fried" just means "flat colors"?
>>
>>8891618
fried is 'hurt my eyes'. how old are you?
>>
Any gen that has no BBC is hurting my brown eyes.
>>
>>8891602
unironically what was LAX thinking
>>
>>8891624
he wanted to show off his big chink cock (and he did)
>>
>>8891624
"fuck we ran out of grant money"
>>
>>8891625
>big
did he?
>>
File: 1770639050134615.jpg (448.3 KB)
448.3 KB
got so distracted by the BBC I uploaded the wrong version!
>>
>>8891629
why does your gen look like herbal detox drink color
>>
>>8891624
LAX really didn't want to train vpred, so something along these lines:
>you kept calling eps grey, FINE, you can choke on your vivid colors, dumb freakazoids
>>
File: 1763974652080915.jpg (393.3 KB)
393.3 KB
>>8891632
your clever 1st world jokes fall on deaf brown ears.
>>
>>8891635
ok ill try again
>>
>>8891629
your gen looks bile
>>
>>8891632
>>8891640
Explain pls
>>
>>8891643
it's really green and bile/vile. bile is green. i pooped green once
>>
???
>>
its a funny joke you're just too dumb to get it
>>
I'm sure everyone laughed at your jokes on the short bus, lil bro
>>
i am sure it is in your homeland
>>
green with envy
>>
I too make japes about the Galenic humors when in company of my fellows at the local Whist club between a glass of brandy and drag of my pipe.
>>
what a nerd
>>
File: Urabe_00013_.png (1.0 MB)
1.0 MB
>>
fried
>>
chicken
>>
you made me go and buy a chicken roll, are you happy?
>>
>>8891697
ethnicity? i want to fantasize a bit
>>
File: 1771594363615234.png (2.9 MB)
2.9 MB
>>
>>8891716
butiful colours
>>
fried
>>
>>8891716
fat_female_lanklet_male?
>>
rice
>>
i like fried rice with shrimps as topping
>>
any hetero gens?
>>
>>8891731
>>8891665
>>
>>8891731
>>8891635
>>
>>8891731
>>8891419
>>
>>8891738
that gen is like 2% female 86% male 10% cum and 65% background
>>
>>8891739
That's not how percentages work.
>>
it's the 2% that matters
>>
>>8891738
same energy https://litter.catbox.moe/t6mkj9.jpg
>>
crazy that we are getting a kino cutting edge sota model like anima for free
>>
>>8891744
it's 2b parameters what the fuck do you mean cutting edge sota model
>>
>>8891746
>muh parameters
utterly mindbroken llmtard
>>
>>8891749
Even if you set parameters aside, in a world of Nano Bananas and GPT-Images that have seamless editing and visual reasoning, it was architecturally outdated even at the time of preview1's release.
>>
so true. i make such butiful hentai with nano banana and gpt image
>>
gpt also can't fart on prey doe
>>
You do not understand what "SOTA" means and are just using the term as a buzzword.
>>
>>8891750
brute forcing trillion parameter models that need entire clusters of h200s to run and the electricity of a small city isn't technological progress tho.
>>
>>8891756
Indeed, but there are editing models that run on consumer-grade hardware so that's hardly an excuse.
>>
use case for /h/?
>>
>>8891757
qwen and flux klein are slop. only fit for jeets and their butiful instagram ai influencers.
>>
>>8891759
Perhaps, but what they do prove is that the tech IS there and CAN run on an average PC. Just because they failed at making good models doesn't mean it's impossible and shouldn't be attempted in our niche.
>>
>>8891756
10-20b is the sweet spot for being able to do a tonne of cool shit by default on consumer hardware tho
>>
Good morning sirs ( • ̀ω•́ )
>>
when everyone is a funny troll poop spammer no one is!!!
>>
>try to generate some figurines on anima
>figure \(medium\), photo \(medium\)
>get nothing but 3DPD or SD1.5esque slop
How the hell do you prompt for this? These are the only two booru tags consistently used for figurines.
>>
>>8891773
just use anime model
>>
>>8891773
>try to generate thing that isn't anime with anime model
>>
https://danbooru.donmai.us/posts/7555875
>this isn't anime according to /hgg/
>>
figurines are /aco/
>>
File: ComfyUI_42158_.png (1021.1 KB)
1021.1 KB
>>8891773
It's just "figure". 7.7K on Gelbooru as opposed to 1.7 for figure \(medium\), most people do not use the full tag.

photo \(medium\) is for photorealism. Figures are too different from real life, closer to a 3D render.
>>
>>8891781
it's a photo of a doll, so no it's not anime
>>
>>8891773
just search for a lora
>>
>>8891785
based screencap chad
>>
>>8891783
Very useful, thanks. I never used Gelbooru since they don't have a usable blacklist system so the thought of using that site's tags instead never crossed my mind.
>>
>>8891789
Anima guy said to prefer Gel tags when possible. He probably used it for training instead of Danbooru, or some mix of both.
>>
>>8891754
can you?
>>
>>8891787
Wow so if I download a lora I can do anything but it will also look like shit?
That's rough.
>>
>>8891798
b-but I put a lot of effort into that gen... how mean
>>
>>8891799
1.5 cutie...
>>
>>8891792
it could just be because gelbooru tags never really changed and this one has a certanin cutoff right?, theirs new pointless, useless tags for danbooru like "alternate breast size", "topless female", "alternate breast size (larger),
>>
File: 1773570295260072.png (2.7 MB)
2.7 MB
>>
Man sometimes I feel like Anima is trolling me
I must be hitting a streak of bad seeds that trash the style adherence
>>
File: 1770215601310146.png (2.9 MB)
2.9 MB
>>
>>8891810
show me her knees
>>
>>8891812
Oh shit, I didn't notice. Never giving the recommended pos/negsloppa another go kek.
>>
at least she has a lower body
>>
>>8891813
idk what i helped you notice. i just have a knee fetish and wanted to see her knees
>>
>>8891816
nta, but I get you, I like slav squatting/squatting riding but 90% of the time its not the frontal knee type of squatting I want to see more, even with knees up.
>>
>>8891816
you're joking, but i get off to shoulder blades
>>
>shoulder blades
/aco/ dogwhistle
>>
kys yourself p*interlyschizo
>>
>>8891819
im not joking. i have a massive thing for legs and knees
>>
https://litter.catbox.moe/6kd7ujsytuzlslue.png
>>
>>8891809
>streak of bad seeds
That's every streak of seeds with anima ^_^
>>
>>8891824
take that yucky thing off her knee
>>
>>8891819
shoulder blades are quite erotic, yes I agree
>>
what settings do you guys use for soft inpainting
>>
>>8891831
I just don't use it anymore
>>
>>8891828
>has knee fetish
>doesn't like bandainds
>>
>>8891828
but what if she gets an infection
>>
>>8891833
yes i don't like wounds
>>
>>8891832
then that question wasn't for you
>>
>>8891824
me on the right
>>
>>8891835
yeah, but it's just fashion thing
>>
>>8891834
ill kiss it better
>>
>>8891838
the only place where bandaids make sense as a 'fashion thing' is on nipples and when used to seal a v*gina (super glued)
>>
>>8891840
explain this then, atheist https://danbooru.donmai.us/posts/1723667
>>
fried
>>
>>8891840
bandaid on nose is a staple for tomboys. bandaids on knees are ultra omega pedo creepy tho
>>
I'm bobbling my head in disagreement
>>
>>8891849
yes, bandaids on knees are very anime
>>
File: file.png (103.2 KB)
103.2 KB
>>8891831
change as you see fit and I am also using the anima-lllite-inpainting-v2 CN model
with all of that now it feels like inpainting on SDXL again
proofies of my lastestest test gen
>https://files.catbox.moe/0qkkg4.png
>>
I'm thinking of baking an oldies style of lora, but images are at a low resolution (1024 x 768), would it hurt to do an upscaler with it when it comes to baking?

i'm thinking of ranch quest, or those heavy enviroment rpg ones/similar to it.
>>
>>8891853
>quest artstyle instead of VI
>>
>>8891853
>ranch quest
eroge if western
>>
>>8891852
>soft inpainting settings closed
>>
>>8891858
just see the metadata on the catbox bro, what are you doing
>>
>>8891855
it's still on paper i'm actually aiming for something like n@gi itsuki, you know the old painterly, watercolor, fantasy ones. however they all have the same low resolution. and honestly all gamecg's in panda.
>>
>>8891570
>>8891742
>>8891574
>>8891593
more
>>
no one wants to see bobbleheads
>>
what if loli but normal head
>>
>>8891869
same reason people do bibici instead of normal penis. it just isn't the same without the bobble
>>
once you go bobble you never go back
>>
what if... bobblehead x BBC
>>
why do you wanna see them poopin?!
>>
>>8891874
it was done before by the bobbleschizo himself
>>
>>8891874
that would just b lolipop with bibici stick (gored body)
>>
no one is safe from the BBC
>>
what the fuck
whenever I use soft inpaint it clearly works, but at the last minute it just becomes a duplicate of the base gen like I didn't even inpaint
>>
>>8891879
cumragUI?
>>
Just forge things ¯\_(ツ)_/¯
>>
>>8891880
forge neo
>>8891882
sounds like it
>>
HOW DO I USE multi diffusion with anima on c*mfyui???????
>b-but why are you using multi diffsion lil bro, that's so last ye-
SHUT UP!!!!!!!! JUST TELL ME!!!!!!
>>
I'm less concerned about you using MultiDiffusion and much more concerned about you using cumrag.
>>
>>8891885
why are you using cumgoy UI
>>
>>8891887
i use it to power my krita
>>
>>8891885
Yeah it wasn't updated for anima, but some anon fixed it and shared here like a month ago.
>>
>>8891890
nice contrast
>>
>>8891891
>>8891890
he's doing the 6 7 with her boobs
>>
>>8891894
can i have box?
>>
>>8891896
thanks.
>>
>>8891895
catbox for this post in photon CSS?
>>
come to the real thread
we have gens
>>
yes please, leave already you fucking immigrants
>>
gens are for browns anyway
>>
>>8891913
I have dual citizenship
>>
>>8891852
Thanks anon, I gave it a few tests and it truly does seem to work like SDXL
At least until it snap backs to the original gen because I'm apparently having the same issue as this anon >>8891879
>>
man why is anima preview 1 and 2 kinda better than base anima
like there are some improvements over them but I feel like styles got worse
>>
>>8891916
yup, it does shit the bed every now and then but that's just anima as a whole at this point, sad times
>>
>>8891918
Could be a Forge quirk, or at least I've never seen it happen on Comfy
feels like the only difference to SDXL is it needs a little higher denoise
>>
>>8891804
I’ll never forgive what they did to otoko no ko (currently trap)
>>
File: ComfyUI_42180_.png (1008.6 KB)
1008.6 KB
>>
Oh yeah, this gen reminds, anime DOES look like shit.
>>
>>8891918
No, it seems that soft inpainting is broken, not sure how it is for that anon but every time I try to gen it works during generation but at the last second it snaps back to the original gen
I did some brief researching and found a thread on reddit of people complaining about this very recently, and it seems like it could be an issue with stability matrix which is what I use because it's useful for organizing shit
Could just be forge neo shitting up the bed
>>
>>8891913
I think only brown people should remain here.
>>
Where little brown bros go, I go.
>>
>this gen reminds
>>
business idea: little brown imoutos
>>
yes with mandatory (smell:4) and (steaming body:3) tags in prompt. also hairy armpits and excessive pubic hair
>>
>>8891911
everyone migrated to /h2h2g/ lil bwo.
>>
>>8891930
I'm no longer rereading what I type before hitting submit, with this being brown thread and all. Just type and send.
>>
>>8891909
uvula box?
>>8891912
nice legs
>>
>>8891939
ty
>>
>>8891943
me in the back
>>
use case for being white?
>>
>>8891948
i get to pretend i'm brown.
>>
https://x.com/Biscuit_QWQ/status/2061884943933051050
the least retarded furry
>>
cucks
>>
>>
cucking is the thinking white mans fetish, no one is cucking indians
>>
ooh she pooped!
>>
>>8891961
my pov
>>
>Hmm...this page doesn’t exist. Try searching for something else.
>>
>>8891961
thought it was that fire force lady from the thumbnail
>>
File: 1768411938321100.png (2.1 MB)
2.1 MB
>>
blud still doing the pov recording cuck thing in June
>>
>>8891988
me holding the phone taking photos for the family album
>>
>>8891989
shit it's june already
wasted another six months

it's time to move on bros, don't let the thread be your life
>>
>>8891989
are you really surprised the retard spamming the same slop since pony has zero creativity?
>>
>>8891989
>June
Nyooo
>>
>>8891992
How do you even manage to waste time here when it's so fucking dead
>>
File: Frogfuckerfrogfucker.jpg (397.9 KB)
397.9 KB
>>
>>8892000
my pov
>>
Which upscalers do you guys use for anima?
>>
i hate anime
>>
we know
>>
File: 1757743962056117.png (2.4 MB)
2.4 MB
>>
i love tranime
>>
>i hate anime so im gonna live in the anime thread for years spamming acoslop
what causes this?
>>
imagine replying to bait. couldn't be me.
>>
i genuinely hate anime fuck you
>>
i genuinely hate you fuck anime
>>
i only watch boruto, jjk and dragon ball
>>
i love aco
>>
>>8892046
your caste? brahmin?
>>
>>8892040
>blud, doesn't know that most of the people in these threads just like big cartoon titties or they are heckin og otakus because they like vocaloids and 2hu
>>
i think 2hu fucking sucks
>>
>>8892051
too soon ...
>>
>>8891575
>>8891635
catbox please
>>
>>8892053
>>8891575 maguro coffee
>>
happy pride sirs
>>
we should post cute boys
>>
Your boxes go into my wallet
>>
>>8892063
>>8892065
kill yourself
>>
futa women are real women
>>
post futa
>>
>https://files.catbox.moe/afec66.png
>>
gay
>>
>it isn't tranny when it's le oldfag
>>
>>
>uncs still on fucking 4channel at 50+
grim..
>>
nice
>>
>>
>>
>>
>>
File: 1girl, s-1544169219.png (946.4 KB)
946.4 KB
>>
1.4 MB
>>
1.9 MB
>>
>>
>>
>>
>>
>>
>>
why is blud posting what is clearly hdg material in hgg
>>
>>
>>8892092
best image itt.
>>
>aco out of nowhere
what caused
>>
File: 1777296917831835.png (2.3 MB)
2.3 MB
>>
(・ิω・ิ)
>>
>aco out of nowhere
what caused
>>
>>8892140
thank you for always posting hentai in this aco shithole
>>
i would be surprised if it would be anime
>>
in perfect /hgg/ we would only generate anime screencap gens and go to bed before eight.
>>
>anime screencap
acojesh dogwhistle
>>
I hate anime too
>>
anime hater thread chuddie
>>
the only anime i like are 2b blender animations
>>
>2b
>anime
>>
speaking of anime, FFVII Remake lora doko
>>
ummmm, gens?
>>
>>8892151
>>8892153
This is an A2 general, chudlet.
>>
>>8892157
>>8892140
>>
>>8892159
no that's not AI enough. we must revert to ayakon.
>>
>>8892160
the slop above it then
>>
File: 1769056727319683.jpg (518.9 KB)
518.9 KB
>>
>>8892169
meandmysonswifebox?
>>
>>8892169
fuck, it's so anime, you have to be a real nipponjin
>>
>>8892169
my pov
>>
>>8892169
him >>8892172 and his son's >>8892175 box?
>>
i'm racist, so i dislike this thread
>>
File: 1775201422908346.png (2.6 MB)
2.6 MB
>>
https://litter.catbox.moe/ebdew757vdxucfv3.png
>>
diddychud ah gen
>>
it's anime
>>
https://litter.catbox.moe/z24oudjuwo2tji3m.png
>>
i love anime
>>
>we love anime artstyle on /hgg/!
>average anime artstyle on /hgg/: https://danbooru.donmai.us/posts?page=1&tags=lewdamone
>>
You're telling me a cuck had all of these POV's?
>>
yeah i did
>>
>>8892193
where
>>
File: 1768900392346285.jpg (350.4 KB)
350.4 KB
>>8892172
>>8892176
https://litter.catbox.moe/dgm8acj3x9j0160w.png
>>
Why are you faggots posting gens
I come here specifically because you don't post so I don't have to see your shit tastes
>>
File: ComfyUI_09743_.png (1.4 MB)
1.4 MB
>>
d-dios mio..
>>
in hindi please
>>
>>8892235
finally, real anime
>>
>>8892235
Box?
>>
>not cross-shaped pupils
Cheap and bad anime impersonation, not anime.
>>
File: ComfyUI_09723_.png (1.4 MB)
1.4 MB
>>8892257
https://files.catbox.moe/am3cli.png
>>
File: 00104-2843536894.png (692.5 KB)
692.5 KB
>>
why are cumfy niggers like this
>>
>>8892266
poking hazard box?
>>
>>8892266
their age?
>>
>>8892269
they're east asian oji-sans in their 40s
>>
https://litter.catbox.moe/763vhh.png
>>
that's a very cute image!
>>
File: 1776454719112467.jpg (434.7 KB)
434.7 KB
>>
>>8892273
stinky
>>
>>8892273
sniffs fingers
>>
>>8892266
sucklebox?
>>
>>8892280
>>8892268
https://litter.catbox.moe/3novqdfbbzzhamhr.png
>>
>>8892217
Based Bong anon.
>>
File: 00152-1494687832.png (919.7 KB)
919.7 KB
>>
>>8892266
is that mori calliope from holomyth gen1?
>>
>>8892193
those images are more anime than any novelai gen I have seen here
>>
>>8892289
is this hentai-foundry lora?
>>
>>
https://x.com/TengaHealthcare/status/2062003839591690267?s=20
>>
>>8892294
more like any gen period
>>
Heck, /ldg/ has more anime posts than this thread
>>
>>8892343
This isn't an anime thread tho, it's a dark-skinned male thread for the most part, the style doesn't really matter.
>>
https://files.catbox.moe/5wkek3.png
>>8891574
yesterday splotch. today splotch.
>>
/aco/
>>
>>8892343
this is the poop and bobble thread
>>
i hate anime
>>
anime? yeah i like thats. there's nothing good to watch though.
>>
imagine watching tranime
>>
i only watch aco
>>
tranima when
>>
jav lora doko
>>
File: 1755846263561820.jpg (422.6 KB)
422.6 KB
>>
Any painterly enjoyers?
>>
>p*interly
>>>/aco/
>>
least vidgens are gone
dreadful things
>>
>>8892366
what caused her arms
>>
>>8892366
that's not even that painterly desu
>>
>>8892372
and that's a good thing
>>
>>8892366
can you add more chromatic_aberration, blurry?
>>
File: 1759124750972484.png (2.6 MB)
2.6 MB
>>
>>8892378
my pov
>>
this place really fucking sucks now
>>
>now
>>
>fucking
>>
>now
>>
:^)
>>
i know! i will save /hgg/ with butiful WAN videos. the more it suddenly drops into realism the better.
>>
>>8892386
>the more it suddenly drops into realism
WAN doesn't do that, that's LTX.
>>
rotoscopa when
>>
>>8892388
mo they both do
>>
>>8892380
Luckily the shitposters are lazy enough that they can't shit up /hdg/ and /hgg/ at the same time so you'll just have to switch between them every once in a while.
>>
anime needs a splash of realism to be good
>>
already has, we need splash in the diaper, actually
>>
File: Comfyui 37082 .webm (801.1 KB)
801.1 KB
>>8892388
Wan is biased towards realism. Don't feel like melting my gpu again so it's a repost, but watch her hands.

Though compared to that, LTX doesn't understand anime at all. It's also far more censored, which probably doesn't help.
>>
>>8892398
the further you push it away from realism/3d the worse it animates. local vidgen sucks balls and saas is censored to shit. vidgens are still 100% meme territory
>>
>>8892399
Thankfully everyone loves 2.5D where it usually works fine
>>
>>8892399
I think WAN shitmixes are at about SD1.5 level. They aren't winning any awards, but at least they're recognizably hentai videos, if extremely shoddy, and not whatever the fuck LTX shits out.
>>
>>8892403
Thing is for img2vid nudity you don't even need a mix, the Wan2.2 base will work fine.
>>
>>8892378
tsunako is really cute
>>
my pov gens?
>>
>>>/g/108972660
That's not too bad, no? Even my boy ayakek learned to gen while you were busy shitposting with negroes.
>>
now you've done it, the schizo will be along soon now to claim it's the worst thing ever
>>
who is lil bud talking to
>>
still looks the exact same as your slop always has ayacuck
>>
also you should just start posting here again. we don't have any good lolcows left u.u
>>
>>8891924
box pls?
>>
>>8892378
Oh yeah almost didn't notice an elf ear fondling gen. Nice.
>>
>>8892423
sure https://litter.catbox.moe/ckycnwf9m01ucsli.png
>>
>>8892414
>>8892415
it's depressing. the cyber bullying stripped all the joy of genning out of him. it got so bad that now all he posts is /e/ on a safe board. the gen itself is inoffensive and bland. good enough with melted fine details. he already knew that though.
>>
>the gen itself is inoffensive and bland
>same gen but add "penis over eyes, dark skin, (huge penis:2), testicles, penis focus, precum"
>OMG CATBOX??? SO GOOD
>>
who is lil brown bro quoting
>>
this is how i imagine the love between hgg's mascot and hdg's mascot :))))))
is this true hgg?
>>
meltie weltie
>>
of course its the fucking cockroach
>>
>>8892434
Looks like modern hollywood where they force every sort of diversity, so you got the black, the fag, and the disabled person posting it.
>>
>>8892434
where are the (smell:3) and (steaming body:2.6) clouds
>>
>>8892437
>so you got the black, the fag, and the disabled person
you're a 3 in 1 bro
>>
beeg mad
>>
>>8892434
where is the ugly fat bastard fucking them both
>>
>>8892442
So your fat brown bastard self-insert is also a faggot who's into trannies? The lore deepens.
>>
>>8892434
more of hims ejaculating_while_penetrated please.
>>
>/hdg/ ruined by iaschizo
>/hgg/ ruined by faggot and bobblehead
retards still refuse to move to /ldg/
>>
immigrants ruined hgg
>>
goyim ruined hgg
>>
https://litter.catbox.moe/equzpr.png
>>
https://x.com/inu__GT/status/2062085540468711673?s=20
>>
>>8892452
>Unable to fetch translation
thanks grok and offtopic poster anon
>>
>>8892447
>/hgg/ ruined by ... bobblehead
explain in great length and detail
>>
>>8892454
when I open a catbox I want to see a majestic stallion pounding a woman with huge breasts, not some funko pop monstrosity
>>
>>8892454
dogshit sloppers always have a raging vendetta against him for being so successful
>>
>>8892455
why are you on the anime porn board if you don't like anime
>>
>anime is when you do dogshit chibi oekaki children colored with the ms paint bucket tool
>>
( ‾ ʖ̫‾)
>>
>>8892455
Finally, someone else who appreciates majestic nature of stallions, maybe we can make /h/ great again.
>>
File: 1776221173685793.png (3.7 MB)
3.7 MB
>>
File: 1763641879321207.png (1.7 MB)
1.7 MB
>>
File: 1752759402914974.png (1.8 MB)
1.8 MB
>>
File: 1752576632157307.png (1.7 MB)
1.7 MB
>>
>>8892456
>successful
>$25 a month
heh.
>>
looks like someone got triggered
>>
File: 1775872948634010.png (3.9 MB)
3.9 MB
>>
>>8892455
the big boy gens go on /hdg/
>>
>>8892465
that's a lot of money in the country of the subhumans obsessed with him
>>
>>8892458
chat is he wrong?
>>
A lot of samefag for a bobbled head
>>
>>8892469
what country would that be saar?
>>
>>8892473
China, Sri Lanka, Pakistan, Bangladesh, Nepal and Myanmar sir
>>
>>8892474
what did my country do?
>>
i am king. you are my bitch.
>>
>>8892480
you are the bastard bitch that stole land from Bharat
>>
mexico
>>
watashi ga kita
>>
starting to think anima might have pruned artist names from the deviantart portion of the dataset
none of the DA exclusive ones seem to work
>>
>>8892489
get rekt tumblr tranny
>>
>>8892486
you can't learn japanese.
>>
ippai icchattaa
>>
File: 1755083533084174.png (2.2 MB)
2.2 MB
>>
>>8892492
use case for learning wapanese on this economy? I would rather learn chinese
>>
watashi ga krita
>>
>>8892496
chinese sounds ugly tho
>>
>>8892496
being able to not rely on trannylators for games/anime/manga, but it has 0 use cases for actually making any kind of money. Japan is going down the shitter rapidly.
>>
>>8892496
Gonna read some light novels I refuse to read in English because translations can never be perfect.
Don't need to make money anymore in this life so the part about economy doesn't bother me:3
>>
>>8892499
>Japan is going down the shitter rapidly.
who isn't?
>>
use case for posting shit gens in this economy?
>>
>>8892502
it furthers global warming thus drowning browns
>>
>>8892502
ask any ai thread about it
>>
>>8892496
fact check AI translations. mog translators.
>>
>>8892503
aren't they just gonna swim to roach lands?
>>
>>8892506
good amount are claimed by the mediterranean sea
>>
why is hdg2 so racism
>>
>>8892507
bahrat alone is over a billion strong. just 1% of our solider enough to deroach the world
>>
>>8892508
Murrican hours.
>>
>>8892508
this how browns bond
>>
brown skin, bondage
>>
https://litter.catbox.moe/ifyx44.png
Prompt stolen from /b/ >>>/b/949922953
>>
>>8892519
Why not post it normally though? It's just body paint on kemonomimi, it's not actual furshit.
>>
>>8892521
too hetero for hgg
>>
>>8892521
Didn't feel like adding brown male to post normally.
>>
If you had to pick one artist and one artist only to use for gens for the rest of your life, who would that be? For me it's got to be Tsunako.
>inb4 some butiful 2.99D tbp slopper
>>
>>8892526
Solipsist.
>>
>>8892501
Faster than others
>>
>>8892526
d*******dles
>>
>>8892526
nyalia
>>
>>8892526
hata4564
>>
>>8892526
firedotinc
>>
>>8892534
oh we're including AI sloppers too? brittle
>>
>Tsunako
anime screenshot tier
>Solipsist.
le painterly /aco/ slop
>d*******dles
hola senior
>nyalia
proto wai slop
>hata4564
butiful 2.99D tbp slopper
>firedotinc
pedo
>brittle
pedo
>>
>>8892536
what about me? :3
>>
>>8892536
>>Tsunako
>anime screenshot tier
Tasteless plebian.
>>
>>8892536
hola amigo *kissu*
>>
>>8892536
you're a pedo too, by the way.
you wouldn't be on 4chan if you weren't.
>>
>>8892538
anime screenshot is a complement. we like anime ova' here.
>>
>>8892538
you can almost be certain that every artist with more than 2k images will be shit. Either they've been at it so long that they've lost touch and draw in some boomer style or they have some low effort style that allows them to shit out images daily
>>
>we
>>
>>8892526
obviously kobuichi
>>
>>8892546
>or they have some low effort style that allows them to shit out images daily
daily images don't have to suck you know
>>
>>8892550
this. just look at firedotinc
>>
>>8892546
Tsunako is honestly an exception, every model that shits itself at old artists who've changed styles over their runtime (most notably right now Anima) nails her and only her, mostly because her style has remained basically unchanged from the beginning to now. Now I understand you were talking about the artist's actual merit and not how they're represented by imagen models, but honestly, if this level of consistency isn't indication of a unicorn who's had it figured out on day 1 and didn't NEED to change, I don't know what is.
>>8892549
That was my runner-up to Tsunako but he very much suffers from the problem I described so I had to take that into consideration. If it was possible to isolate his style from ~Dracu-Riot till now, I would pick him.
>>
wapow
>>
File: ComfyUI_42276_.png (1.2 MB)
1.2 MB
>>8892535
Not like >>8892531 is a real artist either. Also, wasn't that supposed to be bait to make people post gens? If so it failed spectacularly.

>>8892526
Nanameda Kei. I love the look, and he did three somewhat distinct styles I can try and pull out with tag biases so it's cheating a bit.
>>
why would anyone want to post gens here?
>>
>>8892560
Haven't the slightest clue where else to post.
>/hdg/ and altchans are dead
>/d/ and /b/ are abominable hellholes
>/g/ is SFWslop central
>>
>>8892561
x (formerly known as twitter)
pixiv
patreon
>>
>>8892561
>>8892562
chat which one is bait?
>>
>https://files.catbox.moe/io70do.png
>>
nice. very mexican.
>>
>>8892561
/hdg/ is "dead" because it gets gens instead of text shitposts
>>
https://litter.catbox.moe/g1i6zl.jpg
>>
>>8892565
I thought that was Anna Anon from the thumbnail
he's got a lora out for Anima now

>>8892561
Don't forget /edg/ is getting spammed with reposts for over a month straight
>>
File: 1775728752269054.png (2.5 MB)
2.5 MB
>>
anna anon anon, whatever happened to that guy
>>
>>8892574
murica fuck yeah
>>
>>8892565
Decent gen. Not gonna insert it though?
https://litter.catbox.moe/a90syn.png
>>
>>8892579
the horsecock is the best part. you're not supposed to obscure it dumbfuck
>>
>>8892579
I have one with penetration but didn't like it that much
>https://files.catbox.moe/3f7mqy.png
>>
>>8892581
Almost looks like a different style. I like the rendering on the other one better too.
>>
File: 00159-3110859502.png (1.9 MB)
1.9 MB
>>8892582
Yeah I know, I hate the slopmachine on cases like that
>>
>>8892570
me on the bottom
>>
>>8892582
base images were from different artists
>>
>>8892576
There were at least two of us back then, perhaps three. What happens is, you gen so much stuff you get tired of genning.
Although >>8885448 >>>/e/3096881 it works well with vidgen which still feels kinda fresh.
>>
>>8892588
>all the good genners get tired and move on
>p*interlyschizo keeps dumping his garbage eternally
pain
>>
>>8892585
There is no way noodlekek would need a reference image for composition this simple, r-right?..
>>
it was and always will be the same purpose that this place and other ai generals will serve. to make you SUFFER!!!!
SUFFER!!!!!!!!
>>
File: 747576.png (905.1 KB)
905.1 KB
>>8892565
>>8892571
>metadata not found
well that's not helpful
>>
>>8892597
Just roll back to noobai and put artist:deadnooodles in your prompt.
>>
>>8892597
>h*man cock
ewww
>>
>>8892597
it's fucking roxy on a black babydoll holding her legs up with a disembodied horse penis over her stomach, what do you need the metadata for
>>
Do they only teach "on" preposition in Mexico?
>>
>>8892526
pija or daroon for humiliationmaxxing
>>
what do they teach in pakistan?
>>
File: 00028-4048906796.png (3.7 MB)
3.7 MB
cease your shit or i'll send you to brazil
>>
>>8892606
my pov
>>
wouldn't it be an upgrade to their life quality if you did so? geg
>>
>>8892591
you really don't want anyone to post here, don't you?
>>
india > brazil
>>
>>8892609
>lament about good posters not posting any more
>hurrr you don't want anyone to post durrrrr
>>
File: Video_00016.mp4 (155.7 KB)
155.7 KB
>>8892612
be the caused
>>
Anyone got a good upscaler meant for lora baking? I want to revive old stuff but they're ~600x800, 1200,700, etc so I need the least destructive/gunk upscaler for the best accuracy.
>>
do we already have a CN tile model for anima? i'm tired of waiting
>>
>>8892638
>i'm tired of waiting
do it yourself
>>
dead internet theory
>>
https://civitai.red/models/2544636/wai-anima?modelVersionId=2983680
https://civitai.red/models/2544636/wai-anima?modelVersionId=2983680
https://civitai.red/models/2544636/wai-anima?modelVersionId=2983680
>>
>>8892648
local is saved????
>>
anime if real
https://x.com/seanwei001/status/2062178292963443050?s=20
>>
>>8892651
wtf i love anime now?
>>
>>8892648
OMG OMG OMG FINALLY I'M STROKING MY LITTLE CLITTY TO THIS
>>
>>8892660
proofs?
>>
>>8892660
proofs
>>
File: 564.png (1.0 MB)
1.0 MB
>>8892648
its good
>>
>>8892663
bruhhhh. ain't no way this is ai...
>>
HAHAHAHAH
>>
>>8892663
me inside his anus
>>
>>8892666
fake mini anon
>>
>>8892666
i think she's trying to dig you out
>>
>>8892669
don't she dare touch my little brown bro
>>
>>8892648
style loras seem to be a lot weaker with this piece of shit merge
>>
>>8892671
it's working as intended. lmao
>>
>>8892671
>style
>loras
>with wai
are you retarded?
wai doesn't need artist tags or loras. it's perfect as is
>>
>why is my lora weak on this model that came to be through 1000 loras baked alongside it
Truly a mystery
>>
>>8892666
kinky
>>8892671
thats... expected though, fine-tuning/aesthetic tuning does that, its a trade-off between high variance&"low quality" (the base model/low variance&"high quality" (the tuned one)
>>
cucks
>>
bucks
>>
im so horny
>>
proof?
>>
this but hungry
>>
little brown bros.. whatev'ah happened there?
>>
brownout
>>
internet yamero
>>
no
>>
i hate brownime
>>
brownies are dead tho
>>
what happened to them
>>
brahmina when
>>
wrap it up boyos. fill 'er up so us textgods can transcend to 3k next thread.
>>
File: 1761435746721906.png (2.3 MB)
2.3 MB
>>
>>8892729
why is she eating baked little brown bros
>>
never search weird vocaloid songs at night broskis
i can't sleep now
>>
File: 00003-3482343432.png (4.0 MB)
4.0 MB
want sum sexy henti sure why not
>>
File: 00026-2721595703.png (3.9 MB)
3.9 MB
>>
god i want a bvll to rape me...
>>
File: 00106-460980216.jpg (500.3 KB)
500.3 KB
mostly reposts but it should be fine right
>>
File: 00010-1874913656.png (2.4 MB)
2.4 MB
>>
File: 00216-1756430578.png (3.2 MB)
3.2 MB
talk to me let's talk about good things like idk, not being miserable I guess
>>
what caused this
>>
File: 00112-1650322517.png (3.2 MB)
3.2 MB
>>8892741
keeping my mind bussy, feel free to ignore
>>
File: 00294-1519495025.png (2.8 MB)
2.8 MB
>>
File: 00047-3861444533.jpg (829.3 KB)
829.3 KB
>>
File: 00027-1843432451.jpg (894.5 KB)
894.5 KB
I should remake this one on anima
>>
File: 1775599202317372.png (2.9 MB)
2.9 MB
>>
File: 00511.png (446.8 KB)
446.8 KB
that's it, let's all go to sleep while having happy thoughts
>>
File: 00687.png (521.2 KB)
521.2 KB
there
>>
>>8892748
This looks less like pixel art and more like someone upscaled a thumbnail
>>
>>8892748
i hope it's just your usual retardation
>>
>>8892600
I didn't recognize her
honestly TBATE is so much better than Mushoku, didn't deserve that botched animation
>>
>>8892782
kek no
the only amusing part is when he reveals his past life to his parents and his dad gets mad he sucked on his wifes tits
>>
>>8892783
you don't understand, it gets better in vol.5
>>
New thread:
>>8888460
>>8888460
>>8888460
>>
>>8892758
Because the fake pixels are made up of too many real pixels. More than 1.
>>
>>8892791
what the fuck is even going on there now
>>
what's with mentally ill sloppers and ishikei
>>
where do you even see ishikei
>>
>>8892800
please of the understand
they're the only artist they can actually think of by name
>>
had more fun following that guy's story than reading the same 3 shit jokes repeated over and over again by the cripples here
>>
you think we are joking mother fucker?
>>
im jorking
>>
this, we all should start erping
>>
>>8892795
ia schizo spamming the thread and talking to himself in a desperate attempt to make it look active
>>
>>8892813
to me it just looks like he's having a jolly old time over there, unlike you, el constipation from mexico. haven't shit in 10 days, groaning while you write another shitty retarded post
>>
jej
>>
>>8892814
highlighter is also mexican
>>
>>8892814
My hate for brownoids, faggots and other subhumans isn't shared by the mexican, he loves all that. Don't insult bro for the things he haven't done, he has enough other faults you can shit on him for.
>>
>>8892813
He doesn't even gen, just begs for IA requests
>>
>>8892819
what flavor of nigga are you then? fecal?
>>
https://litter.catbox.moe/3fbatmz9fspor67p.png
gm brownbros
>>
>>8892791
why they do not have this indian x bbc shitposter over there
>>
>>8892830
used to
shitposters need an audience so they squat in whatever thread is most alive at the time
>>
>>8892829
>bobbleeta
>>
>>8892829
mindless sex gen with zero thought, love or story behind it. learn from the masterpiecer from the other thread and learn how to give me a reason to care
>>
the sooner /hgg/ dies the sooner we can return to normalcy and reverse the decline

but then again, the split between /hdg/ and /hgg/ did cause a lot of suffering... maybe it should stay that way >:)
>>
decline your weight first how about that
>>
I'd like to go back to /hdg/ but iaschizo is worse than all the shitposters here combined
>>
You post #10 little browns bros wouldn't get it. We were big white sisters pre #10
>>
>>8892859
little brown bros?
>>
waow
>>
i'm jorking to it
>>
we do, in fact, love poop
>>
>>8892869
pairs well with taco :yum:
>>
mmmmmh t/aco/s...
>>
u.u
>>
https://files.catbox.moe/7akg3l.jpg
>>
finally hgg is done for
>>
can someone make a /hgg/ QoS troonjak ACKing xirself
>>
the mexican will finally commit suicide
>>
>>8892869
Me in the back.
>>
controlnet for anima doko?
>>
damn.
>>
>>8892969
>>8892969
>>8892969
>>8892969
>>
>>8892913
>past 9pm
When does it reset? 12AM? 7AM? What if you're nocturnal?
>>
>>8893092
>What if you're nocturnal?
you're untrustworthy.
>>
>>8893094
no you are
>>
>>8888642
damn reup please?

Reply to Thread #8887699


Supported: JPG, PNG, GIF, WebP, WebM, MP4, MP3 (max 4MB)