I made the collage twice. What do I win?

>>101746270
Cancer

>>101746272
N-no.

File: 1714354198245209.jpg (56 KB, 512x512)

56 KB JPG

>>101746287
Don't worry, there's a cure

>>101746249
Catbox an image from your workflow. I will set it up for upscaling for you and catbox it back.
>>101746235
kek at my tiny miku in the middle

>>101746270
>What do I win?
Our gratitude :3

Flux Pro is absolutely NOT better than DALL-E 3 - neutered and censored as it currently is at Image Create - and it can't hold a candle to the first release from last year before all safety faggotry.
Is there any research or evals on prompt upscaling vs "tag prompting"?

>>101746342
nice reference anon
https://youtu.be/3sVBK7K2X6I?t=50

File: ComfyUI_00719_.png (1.06 MB, 1344x768)

1.06 MB PNG

>>101746342
lol

>>101746342
that's insane how good it is at text, like it just get it, what a leap forward everything we've ever known at this point

Still gotta fill up the first thread. Don't get this one nuked.

Why does Flux have that Stable Diffusion vibe? Like it's aura is totally stability pilled. Isn't it really new and stuff? It's giving me the ick.

https://github.com/comfyanonymous/ComfyUI_TensorRT/commit/66f5bf191c8844904adcab436d2ab7c236046aa7

So has anyone tried building the flux wheel or what?

>>101746659
how does that work? it'll make the gen faster?

>>101746235
So premature you fucking retard

File: 48848456565.png (63 KB, 1022x819)

63 KB PNG

>>101746680
Much faster, quieter, more efficient for RTX owners.

>>101746724
oh nice! so... how do I make it work?

>>101746731
Install the extension and follow the instructions on the Github to build the wheel, since Flux is big it may take several hours.

>>101746755
can it build from the fp8 dev instead?

>>101746768
no

>>101746768
No idea, read
>https://github.com/comfyanonymous/ComfyUI_TensorRT/commit/76cdd93680c536a6f5683f60539eeabf27b39775

>>101746774
:(

>>101746778
>Flux needs to be bf16.
but we only have fp16 models?

>>101746774
once you've made a TensorRT file, you can't load it in fp8_e4 like on a regular file right?

>>101746755
Can someone just build it and share the file on torrent?

>>101746788
Flux is bf16

>>101746862
everyone said fp16 since flux came out though kek

File: Screenshot 2024-08-06 192502.png (17 KB, 550x322)

17 KB PNG

I love how retarded and overkill the cooler on the 4080s

>>101746841
Yes, it's different for every GPU and CUDA configuration though

>>101746296
is this bigma

Early bake really took the wind out of it

>>101746968
nah the amerimutts are going to sleep, happened yesterday too

have we got something new that's not flux yet? not a single flux gen is aesthetically appealing

>>101746996
anon thought >>101741569 op was cool

>>101746996
Ai generation isn't aesthetically appealing
Pick up a pencil

official pixart bigma and hunyuan finetune waiting room

>>101746996
pixart team said they'll try to release bigma on september, besides that nothing else seems to be on the horizon.

is there a way to save the workflow on cumfyUI that doesn't force me to type the fucking name each time like a literal fucking nigger?

>>101747073
what do you need to save it for so often? all of your gens have the workflows embedded in them automatically

>>101747084
because it arouses me sexually
is there or not?

File: cui_00580_.png (1.78 MB, 1152x896)

1.78 MB PNG

any good gens lately, anon?

File: (workflow_99999_).png (7 KB, 341x161)

7 KB PNG

>>101747099
edge on windows assigns unique names to each workflow what more could you want you're welcome

>>101747148
you could've just said "no"
I don't mind socializing, tho

File: FD_00462_.png (1.45 MB, 1024x1024)

1.45 MB PNG

>>101746996
quantify "aesthetically appealing" because that is subjective as fuck
>>101747029
picrel also kys

>>101747113
nsfw at present.

I am attempting to defeat the protections they put in place against the booby.

If I want to try flux, comfy is the only way? I have an old version of forge installed and an even older a1111 from the 1.5 days.

I have discovered there's a face-enhance layer somehow.

what is the catbox rule on blurred porn? I didn't blur it, Flux did.

So like at the first levels of the thing, it has a blurry knowledge of porn. But then there is the face write layer. And that happens, and would anyway, but then there is I think more enhance, but then it does clothes write, I think. I don't know how you knock face write and clothes write out of commission

>>101747051
back into my coma for a month, see you all then

>>101745923
Afaik the gen time will be the same with straight to 1024x method. You will only gain time if you upscale it with lighter models like SD 1.5 or XL.

>>101747205
You can post a link on catbox to a woman spreading her asshole if you want to.

>>101747189
i somehow got flux to show a little nipple or areola, previously i was under the impression that it thought females had nothing but skin over the whole breast
don't want to risk someone saying its nsfw though or i'd post

>>101747200
>If I want to try flux, comfy is the only way?
yes, unfortunately

:^) It's a game.

I think it's like

(lowres main gen)
LOOP:
-> enlarge, blur
-> enhance X
-> upscale detail add

X={face, hands, clothes}

idk, something like that.

>>101747200
Trying locally? Yes, currently just comfy
>A1111
2 weeks

Trying online? Check OP. Several options.

>>101747215
https://files.catbox.moe/clnq7v.png

>>101747218
They seem to always build on top of other already trained models with this stuff. Could be wrong.

Apple question: is the M4 coming out, and will it have ML enhancement?

File: file.png (127 KB, 1742x360)

127 KB PNG

kek

File: FD_00508_.png (1015 KB, 1024x1024)

1015 KB PNG

>>101747237
>https://files.catbox.moe/clnq7v.png
Your guidance is too low. Happens to me too on non-porn.

>>101747221
>>101747232
Thanks, comfy it is then.

AHHHHH

I meant to say STEPS. not cfg, if I was unclear.

At step 9, it appears clothing is enhanced. At least maybe. It may depend on other factors.

>>101747251
what did you say anon? lol

>>101747254
Look again bro. The face is painted on top. that's what I want you to see.

>>101747262
>At step 9, it appears clothing is enhanced
of course, it has been censured, they just make the model believe that "naked" means "a human with underwear" or some shit

>>101747264
someone made a reddit-tier joke in response to a serious post and I told them to kill themselves

>>101747237
it really likes to force clothes on and even under clothes when you say see through or mesh or whatever
try this
"wearing nothing facing towards the camera squeezing her breasts" maybe i got lucky though

>>101747272
Oh I see what you mean.

>>101747277
kek

File: FD_00178_.png (1.99 MB, 1024x1024)

1.99 MB PNG

>>101746996
Is this aesthetically appealing?

>>101747276
step 6 has tits still, though this may somehow depend.

step 7 is titty erase mode.

We want to bypass steps.

>>101747297
>We want to bypass steps.
that's not how it work, it will do that process anyway, whether it's at steps 7 or an another number

>>101747262
>I meant to say STEPS. not cfg
but you didn't say cfg? You said layers.
>>101747297
that's not how things work

>>101747293
Yes. However, I suspect anon is trolling

File: FD_00041_.png (1.42 MB, 1024x1024)

1.42 MB PNG

>>101746996
Is this?

File: ComfyUI_00014_.png (1.26 MB, 1024x1024)

1.26 MB PNG

>>101746996
Is this aesthetically appealing?

>>101747306
It's absolutely how Flux works :^)

>>101747335
No, Flux likes to really pump up the hos, and I am investigating nudegate rn, but I will investigate milfgate next.

File: FD_00334_.png (1.76 MB, 1024x1024)

1.76 MB PNG

>>101747321
I know, but I just want an excuse to post some of the gens I really liked, and taking the bait is fun.

File: [flux-dev]_00089_.png (705 KB, 1024x1024)

705 KB PNG

>The Sun
According to FLUX

File: FD_00022_.png (1.59 MB, 1024x1024)

1.59 MB PNG

>>101747335
>a woman
Gay

>>101747254
Also, I strongly suspect this is an error in vae.

>>101747365
Flux hasn't seen a single nipple in its short life.

File: SD_3_00022_.png (1 MB, 768x1344)

1 MB PNG

>>101747372
I It's seen more than SD3.

I'M WRONG!!!!

IT ADDS ALL FIXERS AT ONCE

IT IS A FIXER STEP

BWAHAHAHAHAHAHAHAHAHAHAH

>>101747394
Who asked?

>>101747403
anonie, you don't know enough about how these things work
there are no fixer steps, that's not how any of this works

>>101747413
Focus on fixing your model Lykon it's pathetic at this point.

File: ComfyUI_00964_.png (1.17 MB, 1024x1024)

1.17 MB PNG

>>101747413

>>101747447
is she preggers?

>>101747452
I didn't specify anything of that sort, the model made her pregnant by itself

>>101747474
nips

>>101747372
Flux has a titty eraser step. or a titty eraser substep???

https://files.catbox.moe/isntdy.png

^ Flux knows

That's step 6.

But it erases them in step 7.

>>101747495
those are just the same shitty nipples you can get without skipping any steps
I'm sorry, anon, it just doesn't know what nipples look like.

>RTX 5070 is still gonna be 12 GB
NIGGER

File: FD_00270_.png (1.59 MB, 1024x1600)

1.59 MB PNG

>>101747452
>>101747461
And now I am going to make her pregnant intentionally
>>101747495
I have genned shit loads of nips on Flux what are you even on about.

>>101747506
Could be it has only a vague idea.

But there IS a nipple eraser:
>>101747508
compare to:
>>101747495

>>101747495
>>101747474
>>101747508
Anon...
https://files.catbox.moe/w70kv0.png

File: ComfyUI_00016_.png (1.62 MB, 1024x1024)

1.62 MB PNG

>>101747509
desu flux is working fine on my 1080ti just takes a bit
>Prompt executed in 369.29 seconds

File: ComfyUI_00011_.png (1.33 MB, 1024x1024)

1.33 MB PNG

>>101747529
whats sort of prompt is actually making them topless?
i always get some sort of top

>>101747529
Great job!

Now we need to disable the whoremaker. and muscletoner.

At least it shouldn't be on all the time.

>>101747532
shnell?

FOR THOSE SAYING THERE ARE NO SPECIAL STEPS

look again at this:
https://files.catbox.moe/clnq7v.png

Clearly the face has received enhancement.

>>101747553
YES!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

I love Flux, but what Flux is, it's a system of loras.

any nudity and female celebrity likeness comes from what CLIP knows, which is little

File: ComfyUI_00009_.png (1.33 MB, 1024x1024)

1.33 MB PNG

>>101747562
no dev
there really isn't anything in this world the 1080ti can't do except ray tracing
i mean obviously it has to be at 1080p or below but thats why its called the 1080ti
show me a game that I can't run

>>101747529
Anon is trolling, anon. Report and ignore

File: ComfyUI_00641_.png (885 KB, 1024x1024)

885 KB PNG

We are so back.

>>101747592
so is the guy saying a 1080 is that fast imo. I have seen other strange claims that are impossible.

uhh i guess this is what it thinks a nip is
https://files.catbox.moe/ywb3gk.png

File: FD_00277_.png (1009 KB, 1024x1024)

1009 KB PNG

>>101747545
Prompt is literally just "photo of a naked"
Seed: 264852884050485
I queued up 10 and hit go.
I was exploring latent space
>>101747574
You specified an image of a person so it re-arranged the pixels into a persons face. What exactly do you think this is showing us?

>>101747597
Very clean text

>>101747632
Correct. It generates bee stings instead of nips. You have just discovered what the rest of us discovered 3 days ago.
Here is my 2nd ever gen on Flux
https://files.catbox.moe/g3o9jq.png

Here is my plan. Learn to delete the loras of Flux, and then generate images in 1/3rd the time :^)

side benefit: ho-ification removed

>>101747632
did you try seeing which step adds that?

>>101747619
>369.29 seconds
>that fast
anon what

>>101747682
one legged man using a wolf as support

File: ComfyUI_00021_.png (1.15 MB, 1024x1024)

1.15 MB PNG

>>101747619
>>101747669
i never said it was fast i just said it works

>>101747669
same as my 6950xt.

File: mighy_1080ti.png (1.05 MB, 1919x1029)

1.05 MB PNG

>>101747725
oops meant to screenshot it

>>101747667
https://files.catbox.moe/qsexdz.jpg
Anon you are just wrong. Whatever you think is happening, it's not that.

>>101747736
Sure, I got it a little wrong.

It's doing the enhance face sequence for each step. It has enhance face and basically tone up. It's like a whole set of boosters, that have to get triggered.

>>101747755
no

>>101747736
You do know it is constantly overwriting itself, right?

>>101747755
>>101747765
I am very interested to learn how you think latent diffusion works.

anyone's tested ultimate sd upscale with flux?

>>101747730
Isn't 6950xt much better than a 1080ti specs wise? Is it because no CUDA?

>>101747765
>>101747773
FLUX has noise guidance right? So could the trained guidance be weighted towards certain negatives and making it lean toward censoring?

I have definitely noticed the attention moving around the image during generation (shows up like a hotspot)

>>101747831
It isn't censoring, Anon. There's no censoring step. This is just what it thinks nipples look like, because it was trained to think nipples look like that.
And I use think in the very loosest sense of the word.

>>101747773
Why do you assume they are doing exactly what you think they are doing?

>>101747844
Both can be true. It can both not know what highres tits look like AND be self-censoring.

Again, you know that the image is being edited, and if you adjust the step you can see a moment in time as it's being edited. You know this, right?

File: file.png (41 KB, 572x302)

41 KB PNG

Anons, tf is this?

>>101747844
Also, they have some fun stuff in here.

I found the first (?) Flux easter egg: anus is trained on a bug lmao

>>101747854
This is how latent diffusion works.
https://www.tiktok.com/@pebblepicassos/video/7249998510269975813

>>101747870
tiktok

fuck off, really

>>101747844
I get that there is missing data but the noise still seems to be pulled in a particular direction in a way I don't recall seeing in other censored models

No wonder America is falling behind China, we got a retard here who is Tiktok educated.

Is Flux an auto-Lora?

>>101747335
Tits should be much bigger

>>101747895
It was the first link on google that illustrated my point. Where would you prefer I link from?
Here's leddit
https://www.reddit.com/r/oddlysatisfying/comments/14plfld/portrait_of_a_cat_out_of_rocks/?utm_source=share&utm_medium=web3x&utm_name=web3xcss&utm_term=1&utm_content=share_button
Point is, you feed noise, you say "This is a dog" and it moves around the pixels to what it understands a dog looks like. Then it does that x many times for however many steps you chose.

>>101747938
y-y-you d-disagreed with t-teacher

>>101747898
Is there a way to dump a particular moment, and manually make changes, then continue, and do that again etc?

File: 2024-08-06_00116_.png (2.01 MB, 1280x1280)

2.01 MB PNG

>>101747988
you could just diffuse to certain step, save the img, then change said image do changes whatever/edit the prompt and finally in a different workflow use that image as input vae encode (apply noise or not) and run additional steps

>>101747988
Yes you can stop generating at a particular step and do something else with the latent. KSampler (Advanced) in Comfy will let you do this.

>>101747938
step 5 when requesting 10 steps does not equal step 5 when requesting 20 steps. the magnitude of the step delta is pre-determined as a function of total_steps. you can't take a 10-step gen and "choose" 10 more steps

File: latent.png (53 KB, 539x498)

53 KB PNG

>>101747988
SaveLatent and LoadLatent nodes (in comfy)

>>101748048
What the fuck are you talking about?
You request 50 steps. You feed the latent into a ksampler, start at step 0 and generate until step 20, you take the latent out, do something else, then feed it back into another ksampler where you start at step 20 and end at step 50.

File: 2024-08-06_00119_.png (2.26 MB, 1280x1280)

2.26 MB PNG

>>101748074
ya thats the smart way to do it

>>101748048
Shit I thought you were replying to a different comment. Now I am even more confused.

File: FD_00300_.png (1.16 MB, 1024x1600)

1.16 MB PNG

>>101747976

File: file.png (1.04 MB, 1024x1024)

1.04 MB PNG

how long til we can expect to have a cpp version of flux that can run on CPU ?

>>101748164
>diffusion on the CPU
are you a masochist, anon?

File: ComfyUI_Flux_34.png (1.16 MB, 1344x768)

1.16 MB PNG

How exactly does one boomer proompt efficiently? Do you write a long and descriptive text in t5xxl and a short summary in clip_l? Or do you keep clip_l empty? Or do you just say fuck it and write the same long description in both fields?

bruh, this is fucking insane
with a shitty 3060
slow as fuck, sure
but the quality is bananas
with a 48-64-128Gb GPUs the sky is the limit

>>101748164
If comfy can run flux on cpu only, then would you be interested?

File: 2024-08-06_00132_.png (2.64 MB, 1280x1280)

2.64 MB PNG

>>101748214
ya with these we also will be able to use the promised animation model that Black Forest is working on, can't wait

File: FD_00319_.png (1.17 MB, 1024x1024)

1.17 MB PNG

>>101748164

>>101748226
>promised animation model that Black Forest is working on
O_O

File: file.png (1.05 MB, 1024x1024)

1.05 MB PNG

>>101748182
i'm a poorfag with no money for GPU upgrade, I have 64GB of system ram but only 6 on my graphics card

File: FD_00152_.png (1.46 MB, 1024x1024)

1.46 MB PNG

>>101748213
Literally just describe the image exactly.
>A still frame from a fantasy action movie of A snow warrior. She is wearing dark wolf furs, and a wolf rests on her pauldron. She is strong and dangerous. She is holding a dagger, and stalking her prey. Her eyes are yellow. She has a small amount of dirt on her face.

File: file.png (1.03 MB, 896x1152)

1.03 MB PNG

Prompting is gonna make me a better writer

>>101748213
>>101748247
Also I just combined both clips with a single string, so I am essentially putting the prompt into both t5xxl and clip_l

>>101748240
https://blackforestlabs.ai/up-next/
>no demo
>open-source is, at best, only implied
>no whitepaper
Until we get more than just a few cheesy marketing hooks, I'm not convinced it's not just hype to rope in investors

File: FD_00321_.png (943 KB, 1024x1024)

943 KB PNG

>>101748281

>>101748326
too thick

File: 2024-08-06_00143_.png (2.65 MB, 1920x1080)

2.65 MB PNG

>>101748240
this >>101748301
hype at its best .. well they delivered with FLUX so I give em some slack for now

>>101748301
wew
I mean, sora exists, so it's possible
I was a bit deflated after SDXL, but it's clear that we're on a specs plateau
we just need MORE POWER
with 128/256Gb VRAM, good quality video seems feasible, and the images are going to be insanely good
flux is 12b, then 64b-128b is where things starts to get indistinguishable from reality

File: file.png (1.07 MB, 896x1152)

1.07 MB PNG

>>101748329
My bad g

>>101748353
you're just pulling numbers out of your ass

>>101748353
You are working on the assumption that more always = better. You can oversaturate it. Like over cooking a LoRA

File: FD_00325_.png (1.65 MB, 1024x1024)

1.65 MB PNG

I don't know why, but I never win at this game

>>101748356
yes, but they're reasonable numbers
we're here with 12Gb-24Gb VRAM /12B models, we need way more than that

>>101748383
yes, because that seems to be the case, more parameters = better models, = the need for more VRAM
everything about AI seems very much linearly scalable, at least for the time being

lllyasviel updating Forge again

>>101748403
Does it though? SD3 has more parameters than SDXL and it is dog shit. SD2 has more paramaters than SD1.5 and it is dog shit too.
There more to it than just making number bigger, clearly.

>>101748397
maybe because you're holding it wrong

>>101748403
>>101748430
a big brain doesn't do shit if the data you feed it is shit

Flux does some kind of face replace and muscle tone thing.

File: Screenshot 2024-08-06 224838.png (821 KB, 877x545)

821 KB PNG

>>101748434
They just can't seem to figure it out. Model limitation I guess.

>>101748430
haven't tested SD3 but they clearly fucked up something
SDXL rapes 1.5 when it comes to realism

File: 2024-08-06_00146_.png (2.33 MB, 1920x1080)

2.33 MB PNG

>>101748475
the version of SD3 they gave us is seriously broken and is not even bigger in parameter size than SDXL, the architecture of SD3 supports big models of 8b, but they gave us SD3 medium with 2b parameters and it so gutted that it fails to make simple human anatomy .. they went so far to apologize
>https://the-decoder.com/stability-ai-apologizes-for-disappointing-stable-diffusion-3-promises-much-improved-model-soon/
fail at its best

File: SD3_15624_00164_.png (3.98 MB, 1024x1536)

3.98 MB PNG

>>101748475
>haven't tested SD3
it's a bit broken

>>101748453
perhaps
"holding the backs of playing cards"
"playing cards with their backs to the viewer"
"reverse playing cards"
"face-down playing cards"

Same as before, flux pro and other models:
https://grain-compare-packard-anime.trycloudflare.com/

>>101746316
Yes de3 is still better

What do you mean by prompt upscales, the rewriting de3 does?

I've found that schnell has better prompt adherence than dev when given rewritten prompts that de3 spits out, but de3 is still better at fully vibing with the prompt even when forced to not rewrite, flux is sterile in comparison

>>101748545
What was the consensus on this? Was it a legit case of dataset poisoning to censor and lobotomize this shit for (((safety)))?

>>101748584
SAI is a money laundering company and making AI models is just a front they are not very good at.

>>101748584
probably, men can lie on grass in SD3, women can't, go figure

>>101748611
100+ million they grabbed, well they incubated the team that made SD1 into Black Forest .. so thank god they existed .. who long still? not that long is my guess

File: SD3_15624_00187_.png (3.68 MB, 1024x1536)

3.68 MB PNG

>>101748551
Yeah tried those, they always show the card faces.
>>101748568
DallE3 is only better because it knows art styles better. LoRAs will be our saving grace if they can be trained cheaply.
>>101748584
100% poisoned for "safety". Can't make a woman laying down. Can't draw a nipple. Can't even draw a belly button. Still fucks up hands. Prompt coherency is fucked, it barely listens, you always have to fight against the model to get something good. Unlike flux, which just gives you what you tell it for the most part.
I eventually managed to fenangle it but it was such a chore that it became completely un-fun.

>>101747917
Breasts sizes are as vague as possible with the model, probably yet another thing that was scrambled.

>>101748542
>>101748545
lmao, I did see the news and a few videos about the drama, absolute shitshow
what flux proves, tho, is that the trend seems to be that the more training data, (properly processed, of course), the better the results, but the more processing power it requires
the good news is that, the current structure seems able to handle the load, and by that i mean that we don't seem to depend on any major breakthrough but just finding ways to scale the system we already have, which are probably very wasteful, and will also get optimized along the way

>>101747822
Yes and yes.
CUDA is an insane advantage for Nvidia, mainly because they started way earlier than everyone else.

>>101748368
>>101748480
>>101748661
dope

File: FD_00448_.png (1.51 MB, 1024x1024)

1.51 MB PNG

>>101748542
Don't forget the wonderful meme of "2B is all you need"
>>101748545
Meanwhile on Flux

File: Screenshot 2024-08-06 231952.png (1 MB, 1044x840)

1 MB PNG

Some of the prompts I struggled with on SD3 I ran through flux and got what I was going for on the first gen every time.

>>101748584
>censor and lobotomize this shit for (((safety)))
Yes, and they didn't have enough talent to make up for it with something else like flux writing skills.

File: FD_00001_.png (1.33 MB, 1024x1024)

1.33 MB PNG

>>101748770
Yeah because all the talent went and created Flux. Flux is literally the people who invented latent diffusion.

>>101746316
Yes, DALL-E 3 is better because its dataset includes all the stuff lobotomized from flux.
It's just a different approach :
- Flux makes the model retarded for nudity, known people etc, which has the pro of not needing a moderation endpoint, but results in a worse experience than it could have been even with sfw stuff.
- DALLE understands nudity, knows what porn is, artists, known people, but rewrites your prompts to not show them + relies on a second model to analyse outputs and block them.

>>101748811
Anon, but you're wrong about Flux - it can still generate gore and nudity, and it does need a moderation endpoint which Replicate and fal do have.

>>101748834
Generate a pussy with flux then. We wait.

>>101748811
Oh, and also, I have access to dalle azure endpoints that don't check generated images for NSFW, and I can tell you that DALL-E 3 doesn't really know genitals (very, very badly, just a tiny bit better than Flux). And you can force it to use your prompt directly in most cases, unless it contains nsfw stuff.

File: ComfyUI_00750_.png (1.03 MB, 1344x768)

1.03 MB PNG

Sometimes I have genuine difficulty discerning if these images are just screenshots or actual AI. I gotta look close to find flaws.

File: 2024-08-06_00172_.png (2.98 MB, 1080x1680)

2.98 MB PNG

>>101748811
>DALLE understands nudity, knows what porn is, artists, known people, but rewrites your prompts to not show them + relies on a second model to analyse outputs and block them.
not really, they have a secondary system running that checks the output and gives you doggo when it appears in the gens, the model itself doesn't understand it

>>101748843
It's enough that Flux can generate nipples and gore, it absolutely does need a moderation endpoint.

>>101748811
>>101748843
You are confusing flux pro with flux dev.

>>101748863
The dog is only relevant to bing creator/ms designer, dalle does know tons of styles and characters.

can someone point me to the smaller flux files to dl for the 10-12gb range, fp8 i guess? i got a normal version setup but it uses 15 out of my 16gb vram even with lowvram

>>101748863
>>101748811
ah yea you said that.. I should finish reading before posting

>>101748834
What I mean by "make it retarded" isn't that all images disappeared from the dataset, though they probably started with as few nsfw images as possible, but that the concepts are scrambled.
Nude = underwear and all that.

>>101748084
that would show in the diffusers code which you can look at
at the end of the day you have a noise map that is the seed, you have the steps, that's the amount of noise that needs to be removed (n/steps), the model is rewarded based on the quality of the image at the end of the generation
anything that happens during n/steps is what the model learned to achieve the final image, that means there will be some parts of the image changed over time
that is not to say they did not reward the model additionally for genital free generations which is why it likes to always put underwear on, you can tell SD3 did something like this because it prefers nude colored underwear

Do you all have dedicated PCs for this shit? Like a server specifically for inferencing?

>>101748882
Just use the normal model and change your weight dtype to e3 or e4

File: 2024-08-06_00155_.png (1.08 MB, 1080x1080)

1.08 MB PNG

>>101748876
ya sure.. same tech tho.. right? anyhow if dalle3 was released they had serious problems, cause the dataset definitly knows more than it should

>>101748551
also playing cards face down are a real problem.. did not manage them shown back side yet

>>101748847
I've had access to a less censored version for a while, and while it also still not that good for genital (better than flux, dicks exist for example), it's so much better for nsfw (lingerie, porn poses, etc).
Too bad Dalle without moderation endpoints are the rarest thing right now.

>>101748863
>relies on a second model to analyse outputs and block them

>>101748864
>Flux can generate nipples
not really, they look like candy at best
>and gore
not really

File: FD_00365_.png (2.63 MB, 1024x1536)

2.63 MB PNG

>>101748904
What? This runs on consumer hardware. I just use my gaming rig.

>>101748905
i'm hoping to do imagegen with text gen so the smaller the better, i gotta reserve some space for the text model

>>101748904
big boy PCs are 4090s with 64-128GB of RAM

>>101748904
Running on my older am4 machine yeah, it avoids fucking up my main rig when it uses the available 24GB.

>>101748924
>lingerie, porn poses, etc
When we get controlnets and LoRA Flux will shit on everything else.

I have an rtx 3080ti 16gb. Should i dare to run flux and if so what ui should i use?
I'm illiterate when it comes to ai

>>101748906
>anyhow if dalle3 was released they had serious problems, cause the dataset definitly knows more than it should
It'd be heaven.
NAI leak times 10.

>>101748937
Try schnell for gore, for example. And nipples sometimes do look acceptable. Anyway, you're moving the goalposts, the point still stands: Flux *needs* an NSFW checker for user facing applications, it's not a 100% "safe" model.

File: FD_00369_.png (1.92 MB, 1024x1536)

1.92 MB PNG

>>101748948
Comfy make an FP8 mode that you run just like a normal checkpoint, and included a sample workflow
https://comfyanonymous.github.io/ComfyUI_examples/flux/#simple-to-use-fp8-checkpoint-version

>>101748959
>Should i dare to run flux
You can run it in FP8 (basically unnoticeable quality loss)
>what ui should i use?
Comfy UI is the only UI that currently supports Flux
>I'm illiterate when it comes to ai
Everyone is, even the engineers.

>>101748960
Make that a 1000x, dalle3 leak would be a huge thing

>>101748882
https://huggingface.co/Comfy-Org
fp8 versions no account needed
These contain the vae as well, which is why they're a little bigger than other fp8s

>>101748971
What is fp8?

File: 2024-08-06_00175_.png (2.49 MB, 1024x1680)

2.49 MB PNG

>>101748904
ppl run this on as low as mid tier GPUs
>>101748932
ya ya, its to warm, I didn't read .. peace
>>101748960
it would be, but unlike NAI someone would need to write inference for it (probably would take an autist a day or two) and you would have MS and OpenAI on your ass, not a garage corp like NAI .. I wonder if the one who stole the NAI model ever got legal repercussions

>>101748972
Yeah, the biggest jump in local ever, equivalent to a recent claude model leaking for LLMs.
Will not happen, they keep that as safe as they can, but a man can dream...

>>101748882
Whats the matter if its consuming 15gb?
You have 16 for a reason

File: fp8vsfp16vsPro.jpg (853 KB, 2971x1092)

853 KB JPG

>>101748971
>basically unnoticeable quality loss
For comparison sake
FP8, FP16, Pro

>>101748992
>you would have MS and OpenAI on your ass
Once it's out it's out, even if people lay low for months and share this on p2p only, they won't be able to do fuck all.

File: 2024-08-06_00178_.png (2.65 MB, 1024x1680)

2.65 MB PNG

>>101748986
floating point 8 .. the weights in a model are stored in floating point numbers of a certain bit length, the default for flux is 16bit precision, which makes the model 24gb in size, which will barely fit on a 3090/4090 (there is some overhead, so its less than 24gb) .. but reducing the length artificially to 8bit you reduce the size of the model (~12GB) and can fit it in less VRAM, but lose some quality (there were comparisons, its not much of a loss)

>>101748967
literally me having never used it
>>101748981
>its even single file
sweet, thanks for the spoon feed anons i appreciate it. 16gb isnt much to play with for vram but a text model is going to need at least some space else it'll be too slow

>>101749013
The reflection effects on pro are good.

>>101749013
this >>101749020
someone steal PRO pls

File: FD_00370_.png (1.23 MB, 1024x1024)

1.23 MB PNG

>>101749020
I know, pro is fucking sick, so I don't think they need to be worried about community fine tunes on Dev.

>>101749016
Holy shit i just bought this graphics card and i'm already a vramlet

>>101749007
i want to be running an rp at the same time but occasionally inject images of what it thinks the surroundings are (based on my lorebooks and stuff). i can adjust the amount of layers the chat model uses, but to gen an image it'll just crash if it doesn't have the amount of vram it wants. sd already works fine, but i want to plug flux in next

>>101749048
get a second gpu for rp

>>101748986
fp8bp

>>101748561
Added more configuration to pro from replicate - step count and interval (pro-only feature, apparently influences variance by the cost of prompt following, idk how its different from guidance)

>>101749060
you need like 6 of them for this hobby. at least i can shift around some things and run both at the same time (hopefully with these smaller files) like i do with sd

>>101749048
Thats nice. What llm are you using?

File: FD_00374_.png (1.63 MB, 1024x1024)

1.63 MB PNG

>>101749048
In that case use schnell, it gens in 4 steps so your gpu will be whored less.
4080 and 32GB RAM for the spill over and my ram is sitting at 26GB used. GPU 100%.
>>101749060
This, pretty much. It's a phat model.

Anyone has a workflow for tensorRT for flux?

Flux is the nazi chud model

>>101749138
prompt?

File: FD_00002_.png (2.46 MB, 1024x1536)

2.46 MB PNG

>>101749122
Damn right.

File: flux1-schnell_fp8_unet_00011_.jpg (333 KB, 768x768)

333 KB JPG

>>101748243
I don't get it. That's enough to run it already.

>>101749161
>This image is a high quality 2011 screenshot from a TV anime which is visually striking for its rich pastel color pallet soft line art and three tone shading creating a soft and inviting visual design. The image shows the two main characters in a locker room, one of the girls is pulling up a pair of tiny pink cute panties over her wide hips and plump bottom causing them to ride up her butt crack slightly, the other is pulling her shirt over her head, undressing, The view of the scene is partially obscured by black lines giving the impression that the camera is behind a grill or vent. The subtitles read, "I think I heard something from that locker."

>>101749209
thank you, appriciate it.

File: FD_00375_.png (1.44 MB, 1024x1024)

1.44 MB PNG

>>101749208
how long per 768x on 6gb?

>>101749091
l2 miqu has been my main for a while, but command-r+ 103b and the new mistral large 123b are very good without any tuning. i've been spending the most time lately with the new llama 3.1 70b instruct and non-instruct but i haven't made up my mind about it yet, sometimes i think its not any smarter than l2 was and writes worse for rp.

>In that case use schnell
now that i saw that name i remember reading thats the smaller one. good call. i'm not even looking for characters, just broader stuff like a forest in winter, to go along with the scene. so i think i'd be fine with whatever the literal worst version of flux is as long as i can keep the vram usage down

File: FD_00379_.png (1.73 MB, 1024x1024)

1.73 MB PNG

>>101749230
Schnell isn't smaller, it's the same size, but it will gen an image in 4 steps as opposed to 20 on dev, and they will be good enough for your purposes I think, but not as good a dev.

>>101749229
100-120 secs

>>101749256
4 steps just isn't enough for text and fine details, the model has to infer/assume too much per step.

File: 487.png (1.62 MB, 1152x824)

1.62 MB PNG

>>101749122
lets change that then

File: FD_00373_.png (1.76 MB, 1024x1024)

1.76 MB PNG

>>101749257
That's actually pretty decent. How much RAM? 32GB I assume?
>>101749261
I don't think hes embedding text for his images, he is generating scenes from his rp. Text doesn't seem like a requirement, nor does fine detail.

File: emerald cheese.jpg (345 KB, 1024x1024)

345 KB JPG

prompt: emerald cheese

File: 2024-08-06_00190_.png (2.77 MB, 1024x1680)

2.77 MB PNG

>>101749290
tasty! I want it

File: 2024-08-06-095640__0.png (1.06 MB, 1024x1024)

1.06 MB PNG

Took about 12 minutes

>>101749285
16GB. It's limited to fp8 and Schnell model though. fp16 will rape my ssd pagefile drive, and dev model you will need more patience, around 20 minutes for 20 steps of 1024x gen.

>>101749256
why is it 4 steps specifically? steps was always an option before wasnt it? (been out of imggen for over a year)

File: ComfyUI_00235_.png (2.42 MB, 1600x1280)

2.42 MB PNG

>>101749122
>Flux is the nazi chud model

>>101747237
>https://files.catbox.moe/clnq7v.png

honestly i thought you were retarded but now that im trying it myself i think you're on to something, it naturally gravitates towards clothed when you put up the step count even though there is no mention of clothes

>>101749339
It's just the way schnell was made.
It's able to output coherent stuff with just 4 steps
No idea how they did it.

File: 2024-08-06_00197_.png (2.79 MB, 1024x1680)

2.79 MB PNG

damnit flux, DEATH is card 13 .. not 12

>>101749342
It gravitates towards clothes because nudity is extremely limited in the dataset, I would bet that nudity is counted in the thousands in the dataset

>>101749377
and their caption model probably avoids describing the nudity, you see it happen with open caption models like CogVLM

File: 20240806_142104-gen-99574(...).png (916 KB, 896x1152)

916 KB PNG

>>101748355

>>101749348
You have a teacher model (50 steps) and a student model (4 steps), you train the student model against the outputs of the teach model.

>>101749398
I mean there are some tests you can do when the dataset isn't properly captioned, for example, truly uncensored but poorly captioned will show you genitals if you get the right angles (like butt from behind and below on a woman). But when you do that you will see a thong/underwear.

File: FD_00013_.png (1.3 MB, 1024x1024)

1.3 MB PNG

>>101749328
>around 20 minutes for 20 steps of 1024x gen
This is how long it takes me on a 4080 with 32GB RAM of I run in fp16.
>>101749339
It's a "turbo" model. SD has one too. SD has a cascade model which does near real time generation. We have lots of toys since last year, multiple competing models too, and all of them have been made obsolete by Flux.
>>101749342
>>101749377
This is 50 steps.
https://files.catbox.moe/stxrq5.png
>>101749398
It's probably closer to this. Like how copilot sees blurred faces.

File: Schnell_steps.jpg (2.36 MB, 6912x665)

2.36 MB JPG

>>101749348
4 is the minimum steps with decent visual result. Though the manual recommends to use 4-8 steps but you still can use more steps apparently.

>>101749400
Oh, clever.

>>101749440
it's why it's a one-way trip though because you essentially kill the student model for learning new knowledge, so to train the student you have to retrain it back to 50 step because high steps is a crucial part of teaching diffusion models

File: GUS4tALaYAArF56.jpg (323 KB, 2486x1307)

323 KB JPG

training cunny lora
>43.15it/s
wish me luck bros

>>101749437
i still dont know why this is different than dev though. is 'steps' not the same as in auto1111 selecting 30 vs 40 steps? is schnell the same size in vram/ram as dev or smaller?

>>101749472
staber guy strikes again

>>101749475
A model works by denoising n/steps
Schnell is trained to remove 25% noise per step
Dev is trained to remove 2-5% noise per step
That's the difference

>>101749475
schnell was forced to learn how to generate the same images as dev in fewer steps, this necessarily causes it to have issues with some things since each step is a bigger change to the image.

File: 2024-08-06_00205_.png (2.58 MB, 1024x1680)

2.58 MB PNG

>>101749475
>schnell the same size in vram/ram as dev or smaller?
same size
>>101749475
>is 'steps' not the same as in auto1111 selecting 30 vs 40 steps?
yea, but depending on sampler and model type more steps doesnt mean better output, if the sampler isnt something like ancestral it just will converge to an output at certainpoint and more steps more or less just shifts a few pixels

Btw anons, please don't forget that:
flux.1 pro is the original
flux.1 dev is a normal distill from pro
flux.1 schnell is a 4-step distill from pro (not from dev, obviously)

File: FD_00012_.png (1.32 MB, 1024x1024)

1.32 MB PNG

>>101749321
You bent your GPU pins!
>>101749437
I was running 20 steps constantly before I realised what schnell was. I thought it was the smaller model, didn't realise it was a turbo model. So all my first gens are on schnell, and I was happy with the results, but didn't realise why I was struggling so hard with the text.
>>101749472
>43.15it/s
>36 hours
On runpod this will cost about $50. Not sure I am willing to pay that just yet. Interested in your results, even though I hate cunny.

>>101749513
pro leak when?

>>101749489
so the dev one just does more work per step/pass. why cant they just say that lel

>>101749526
they do, if you know how to read

In the next few years it'll be a miracle if we get any woman at all in the released base models, clothed or not.

If you want some non-autosaging bread then it's right here...
>>101749563
>>101749563
>>101749563

>>101749513
pro is just dev + secret prompt rewriting (oh i mean, 'interval') to farm $$$
dev is just a poorly optimized handicapped version to make schnell look good
schnell is the real one

>>101749604
Yamada is that you?

>>101749395
prompt please

File: FD_00245_.png (1.14 MB, 1024x1024)

1.14 MB PNG

>>101749603

>>101749628
>emerald berries

>>101747863
did you notice any difference when trying different values?