#🏞|general-with-images

1 messages · Page 25 of 1

glossy herald
dense tapir
oblique nimbus
#

I still haven't been able to gi that to work on my pc

#

git^

dense tapir
#

Why not?

#

what does it do or not do?

oblique nimbus
#

don't know and now my graphics card keeps crashing

dense tapir
#

laptop?

oblique nimbus
#

nope

dense tapir
#

well, then it shouldn't be from heat having killed it

oblique nimbus
#

no

dense tapir
oblique nimbus
#

and now my wife needed to borrow one of my ram chips

#

so it's slower now

dense tapir
#

ouch

#

I just upgraded my ram about 2 weeks back to 48 gigs

oblique nimbus
#

cant play my favorite games

dense tapir
#

slower but way more comfort

#

I am still saving until about Oct for a 4080

oblique nimbus
dense tapir
glossy herald
dense tapir
#

4090 is the best and 24gb but 1600+

oblique nimbus
dense tapir
glossy herald
oblique nimbus
dense tapir
fresh hound
#

really nice job guys

dense tapir
fresh hound
#

whats your hires. fix settings to not get any weird glitches/doubles at that res please general?

dense tapir
#

oops, I closed it down

#

I don't use hiresfix it is aweful and really painfully slow

fresh hound
#

oh right - didn't know there was alternatives tbh

dense tapir
#

@smoky oak gave me the juice. Get the extension that is actually a script called ultimate SD upscaler

#

use img2img and set denoise to about 0.2-0.26 cfg 7, 50 steps DDIM

dense tapir
#

use esr 4x and set it to what you want

#

it is bloody fast

#

not perfect but better than most

#

@fresh hound

fresh hound
#

that so fucking cool

#

Bruce Willis as Hank from Breaking bad

#

what sampling method do you use btw General?

hardy tulip
#

all my images end up having this weird texture, i wonder if it's the model

fresh hound
#

bruce was good but anyone else i get this

#

damn so close

hardy tulip
#

walter's head is a bit more round

#

and beard is more grown

oblique nimbus
hardy tulip
oblique nimbus
#

I think I posted this already but

hardy tulip
#

those are cool but they are missing something i feel

hardy tulip
#

the glowing one

oblique nimbus
#

different prompt

glossy herald
fresh hound
#

how do I get safetensors to work in 1111?

#

saw online u just past it in the folder and load like a ckpt model but it wont work :C

glossy herald
#

well, maybe do a git pull if you are on a really older version

#

but it worked out of the box for me, putting .safetensors in the models folder showed them in the dropdown menu in the UI as usual

fresh hound
#

looks like I am up to date

glossy herald
#

not sure there

fresh hound
#

ah god damn it I know why now

glossy herald
#

that's quite uncanny

hardy tulip
#

does anyone know why my images have this weird ass texture?

oblique nimbus
#

what do they call that part of a game screen that gives you information on how much life hunger and toolbar? I need it for a negative prompt

languid oriole
#

just look at this master piece

oblique nimbus
celest granite
#

Test

oblique nimbus
#

I get them every time there's a video game title in the prompts. so I needed to add that to the negative prompts

#

and I couldn't remember the word for them

trim jacinth
#

should the preprocessor be on? I'm going to make a sketch of the picture it will look like first?

glossy herald
#

you have a picture of what for now ? a real building ?

trim jacinth
#

yes

#

im trying learn how it works

glossy herald
#

put the preprocessor "dept" on, and click on "Preview annotator result"

#

this will show you what controlnet would take from that picture

#

for buildings, mlsd is also really good

#

that preview button is essential in lots of modes

#

it lets you see what will get extracted, what contours or depth, ...

#

let's take an example

lyric imp
#

I found infinite backgrounds for retro/synthwave youtube mixes xD

trim jacinth
glossy herald
#

you don't download it at that step, but you'll be able to do it soon

#

now that you have your depth map, if it looks alright, you select the same model, "depth"

#

and you prompt what you need, like "photo of a house"

#

here is what I got as depth map

lyric imp
#

I found depth to be better than other mask extraction methods

glossy herald
#

and using those controlnet settings :

#

I get those results

glossy herald
lyric imp
#

although body pose is pretty neat for humans

trim jacinth
glossy herald
#

my personal favorites are canny, and scribble

glossy herald
#

it will use the depth map it made for it anyway

lyric imp
glossy herald
#

it won't reprocess it each time, don't worry

glossy herald
#

well with better prompt than "a house" lol

#

in particular, prompting the building materials is quite strong

lyric imp
#

I wonder how depthmap is processed anyway, because others like canny are basically just binary 1 and 0, but depth looks more like 0 - 255

glossy herald
#

I'm not sure. the model is called Midas and has been around for quite some time

#

deforum uses it too

lyric imp
#

ooh thanks i will take a quick look

glossy herald
#

deforum is a little hard to use, but anybody interested in animations using SD should take a look at it, fantastic results

trim jacinth
#

should the preprocessor be on when rendering the image?

glossy herald
#

if you are using a normal image as controlnet input : yes
if you are using a depth image : no

#

you can download the depth image, it appears as one of the outputs when you make a picture

#

and use that one instead of your initial image

#

in that case, remove the preprocessor

lyric imp
#

I wonder if deforum would work with body pose, IE taking a video of a person and for each frame get the pose then run SD on the pose

#

that would be sick

glossy herald
#

not yet

#

this is whiterun 🙂

lyric imp
#

but theoretically it could be possible without deforum as body pose is all that is needed

lyric imp
trim jacinth
lyric imp
glossy herald
#

nice !

trim jacinth
#

thanks!

glossy herald
lyric imp
#

I saw a couple of vids where the input was someone dancing and the output was a drawing of it, pretty sure they used on of the preprocessors to extract the mask for each frame, but part of me does not want to believe that the person did that manually frame by frame xD

oblique nimbus
glossy herald
#

you extract all the frames using ffmpeg

#

then process all of those through automatic using the API currently, no script does it yet that I've seen

#

and you build the video back up in ffmpeg

#

use the same seed on each pic for more stability

#

some dejitter if you can afford it too

lyric imp
#

true, that would work

glossy herald
lyric imp
#

ffmpeg is underrated like crazy, its insane what it can do while being open source

lyric imp
glossy herald
#

I'm moving from controlnet, and going into the "make 1920x1080 in one pass"

#

longer for sure

lyric imp
#

this makes me wonder if at some point the AI will surpass conventional ray tracing in terms of computation

oblique nimbus
glossy herald
lyric imp
glossy herald
#

not good though

lyric imp
glossy herald
#

even going as high as 5120x1440

#

gtg for now though, but happy to have met you, I hope I'll catch up with you again !

lyric imp
#

oof, are you on an A100 or something xD?

glossy herald
#

3090TI

lyric imp
#

yes, see you ! glad to meet you as well 🙂

glossy herald
lyric imp
#

mine is a 4080 but cant handle it

glossy herald
#

I need better prompt for that

#

the 40xx series is nice on some sides but yeah.. you have 12GB on that, right ?

lyric imp
#

yep only 12

glossy herald
#

yeah, even 3080TI has 16 in some case

#

that's bull crap

lyric imp
#

it is 😦

glossy herald
#

well, catch you later !

lyric imp
#

take care, see you

trim jacinth
#

and which control model best for creating human and animal images?

midnight bison
#

Hey Everyone I Really Very Urgently Need Help With My App.
Is There Anyone Just Anyone Who Could Help Me With Integrating Stable Diffusion Or Any API That Can Simply Take 3-5 Images Of Users And Create It's AI Avatar, Sort Of Like Lensa.
Please Anyone Just Any One Help Me With It It's Extremely Really Very Important.

#

Even If Anyone One Of You Know A Person Who Could Help Me Please Connect Me To Them Just Please Help Me Guys.

wispy nest
#

Why Every Word Starts With Capital Letter, It's Painful To Read

lyric imp
#

sorry to say but you are very late to the game @midnight bison

midnight bison
#

I Want It I Just Want It, My Business Model Is Very Different From Them.

#

Anyone Who Could Help Me I Am Ready To Offer 1% Of My Company To Them.

lyric imp
#

you're planning on using SD APIs? How would that scale up to > 1000 of users at the same time

#

what cloud infrastructure do you plan on using?

fallen swan
#

a guy with black hair, a blue sweatshirt, sits on a throne, a crown on the back of the throne, a smile on his face

midnight bison
#

What Did Lensa DO???

#

Which API Did They Use

fallen swan
#

how to create a drawing ?

midnight bison
#

Any API Just Anything.

#

Whatever It Takes To Make It.

lyric imp
#

so you want to challenge a 25 M$ business, okay, then I assume you must have at least a couple of Millions to invest in your startup

fresh hound
lyric imp
fresh hound
fleet goblet
#

It seems I needed a WSL update

#

And now images seem to work fine without the token limit most of the time.

#

Sometimes however, you get images like this.

#

But when it works, it works.

fresh hound
#

id have thought it makes better lables for the bottles tbh

#

im sure i've seen better results for that part

fleet goblet
stray parrot
#

Has anyone ever get experienced get trojan while doing preprocessing images?

fresh hound
lyric imp
stray parrot
austere pilot
#

@reef frigate this is what I am getting but I don't have the obj file

#

I am trying again cuz it could also be cuz collab was diconnected

#

I just see this

#

it gives here yellow lines as well

wispy nest
#

Vintage 90's anime style Jesus cristo with hair black

crystal deltaBOT
#
FAQ: How do I generate images? Is there a bot on the server?

Currently, there is no bot on the server that generates images. However, there are plenty of other ways such as the official https://beta.dreamstudio.ai/ website or by running Stable Diffusion locally using your own hardware! Check out #1080946152318443610 for more details! You can also stop by #1025467151206854736 for any issues you experience while using DreamStudio or #🤝|tech-support for any problems you encounter while installing it locally!

stark vine
fresh hound
humble badger
#

Can i generate images here ??

#

Can anyone help

spring ledge
#

#⛰️

humble badger
#

#💋

#

This didn't work 😂

spring ledge
#

#🏞️

fresh hound
river lichen
#

/retrowave, synthwave style, palm in the night in the rain,the Last of Us , landscape portrait, 8k resolution concept art portrait by Greg Rutkowski, Artgerm, WLOP, Alphonse Mucha dynamic lighting hyperdetailed intricately detailed Splash art trending on Artstation triadic colors Unreal Engine 5 volumetric lighting, gothic clothing

spare lantern
# stray parrot So what should i do? delete all images that i just download? I mean i only downl...

Some others were discussing this in #🤝|tech-support, I got these same issues today -- not just torch-deepdanbooru, but also Promptgen. It seems like Windows Update changed something and Windows Defender is picking up signatures within pickletensors and bins as being malicious. It may be nothing but it's probably not something to ignore, best bet is to find a safetensor conversion of the model.

smoky oak
#

Ummm

#

That was weird-

dry crow
#

Yea got that a few times

austere pilot
dry crow
#

App restart helps

smoky oak
#

Dang, never seen that before

#

And I am CHRONICALLY on discord lol

#

@dry crow I figured out how to absolutely absurd ultimate SD Upscales lol

#

Like, 100,000,000+ pixel Upscales

#

On just an 8GB GPU

dry crow
smoky oak
#

Yessir

#

Upscaled to 15360x6528

#

Bit of an improvement lol

dry crow
#

What was the source res?

smoky oak
smoky oak
#

I went from 2560x1080, then I did a 3x upscale, took that result, and did a 2x upscale on that

dry crow
#

Ahh nice

dry crow
smoky oak
#

Ultimate SD Upscale, like I said lol

dry crow
#

I mean which Upscaler xD

#

The .pth file

#

Or ldsr

#

Or esrgan

smoky oak
#

For 1x to 3x, I used R-ESRGAN+4 whatever

#

And then from 3x to 6x, I used no upscaler

#

Which made a slightly soft image, which I then ran through gigapixel at 100% scale (no upscale) and it was able to extract detail out of the slightly blurred inage

dry crow
#

Ah okay you used gigapixel for the last step

smoky oak
#

Yeah, all it did was de-blur the image

#

All the upscaling was in SD. I just used giga pixel as a pixel filter

#

Going that high res doesn't work in SD. It tries to do almost 800 upscale tiles, and it crashes

dry crow
#

800 tiles? Of 512?

smoky oak
#

No, that's now how that works

dry crow
#

Did you tried with 768 tile Size and esrgan 4x ?

smoky oak
#

It's complicated

#

Also, you have to use a prompt to get results as crisp at this res

dry crow
#

Okay, now i get it.

#

I need to try the ultimate upscale script tomorrow again.
But i can get 6k images with the normal sd upscale script and esrgan based upscalers.
12k with Extras tab

smoky oak
#

@dry crow the only thing to think about is that the upscaling in the extra tab just add more pixels with no new info, where Ultimate SD generates the new pixels

dry crow
smoky oak
#

It's like giga pixel vs high res fix

smoky oak
#

Just crisper lines. Doesn't work with higher details tho, but that's not a problem for non realistic stuff

dense tapir
#

I love and hate the new UniPC sampler

#

when it works great but most times - CUDA error: CUBLAS_STATUS_EXECUTION_FAILED when calling cublasGemmEx( handle, opa, opb, m, n, k, &falpha, a, CUDA_R_16F, lda, b, CUDA_R_16F, ldb, &fbeta, c, CUDA_R_16F, ldc, CUDA_R_32F, CUBLAS_GEMM_DFALT_TENSOR_OP)

#

it is funny because if you try it again it might work. A bit strange that one.

#

Adaptive takes 10 times longer from 36-45s (depending on sampler) to 4:48-5m

smoky oak
#

So, I should be able to get the new uniPC sampler by updating my a1111, right?

dense tapir
#

yes

#

bottom of the drop down list

#

I like it and it is DDIM

#

has a few quirks to it which I believe is in the auto implementation

#

adding one lora I went from 40s to 49s to gen. each lora makes it worse and worse too

#

@smoky oak

smoky oak
dense tapir
#

VoltaML is tensorrt based for speed and Pascal cards don't have them.

smoky oak
#

Aha, I see

#

Is voltaML a form of GUI I can run?

dense tapir
#

No

smoky oak
#

Ah, alright

dense tapir
#

They had a big presser and a run up to it a few months back

#

they were saying to change one line of code could get you 2.5x speed increase. Yeah, if tensor cores are on your card

smoky oak
#

Is there a way to run this sort of thing on my system?

dense tapir
#

of course

smoky oak
#

So what exactly would the workflow be like to get this to function for stable diffusion?

dense tapir
#

read the link. I don't see people switching to Pytorch 2 if they are using VoltaML

smoky oak
#

Ah, alright, I'll take a look in just a moment

dense tapir
#

I will say if you pull it off 5-6x the speed

#

in SD your 10.5 would become about 40

#

30-40

#

more tensor cores on your card the faster it goes.

#

A100 is fall over speed

#

0.2s to gen an image

#

I was so hyped but then it got closer and the truth came out that we need tensorcores

#

thing is you have to use the specific (as the chap said to me) versions or you lose speed. This is what I really hate about all this py stuff as it is so connected to one version for all of it.

#

and they do mean == not => or <=

#

I just asked for you so I will let you know when they reply

stark vine
#

can someone post some photorealistic mj v5 pix

#

I cant find a showcase in their discord

celest geyser
#

(this is SD, not MJ, just saw the message above)

stark vine
#

yeh I wanna see mj

fiery lichen
#

mj v4 simple prompt @stark vine

stark vine
#

nah not v4, I wanna see dat v5

#

the so-touted insane model lol

fleet goblet
fiery lichen
# stark vine nah not v4, I wanna see dat v5

i am talking about the simple prompt thing. This is from pick-a-pic that was recommended for a demo on general chat by @vague oasis but longer prompt compared to the mj v4 one above

stark vine
fleet goblet
#

Looks like I don't need 75 tokens anymore on the OneAPI version

#

Intel arc working swimmingly now.

#

1.65IT/s on 768x768, EulerA.

#

Of course however

#

Just like CPU inference, there are glitched outputs from time to time.

stark vine
#

hey thanks for supporting diversity in the gpu world

#

that will help bring prices down for other companies, hopefully

fiery lichen
stark vine
#

Aaah I see it now

#

yes I have to admit it is 100% superior in creating small details

fiery lichen
#

but we can do that on 2.1 or even realistic vision 1.5 no problems

stark vine
#

sadly, SD still sucks dick for intricate details, in my case, I am desperate to get good skin textures

#

thats why I keep begging to PLEASE someone make a 1024x1024 model

#

so SD is definitely lagging behind in that aspect

fiery lichen
#

that was upscaled...if you saw what Sytan up there did....

fleet goblet
fiery lichen
stark vine
#

thats the thing G

#

2.1 was supposed to be the bomb

#

but it got nerfed due to the controversies lawsuits etc

#

it is definitely capable of something like v5 can do, if it didnt suck ass at everything else

#

thats why everyone shifted focus to 1.5

fiery lichen
#

for sfw portraits, I have no problems on 2.1

stark vine
#

I would love to see what v5 and dreambooth could do, sadly, that will never happen because mj is the Apple of AI software lol.

stark vine
fiery lichen
#

although i don't make enough portraits as my main work are objects and buildings

#

can i send and not make it public?

stark vine
#

yeh

#

afaik 2.x were trained on stock photos only

#

so good luck generating anything mildly interesting

fleet goblet
dense tapir
smoky oak
#

BACK

#

@dense tapirSorry, I am looking at what you said now

swift harbor
#

Hi, a quick question - when I compare the results I get from the sample prompt images from CivitAI, mine keep showing washed out, fuzzy, and low contrast in comparison when using the same model, same prompt and same extensions. What could this be? I would appreciate any help. I am attaching an example image to show what I mean. The reference is the original image from CivitAI, and the one below is what I get.

smoky oak
#

I have 2 installations of SD right now lol

dense tapir
smoky oak
#

so I could remake the pytorch 2 version into what they need

#

I have 152 tensor cores

dense tapir
#

naw, as far as I am concerned Pytorch2 is doa

smoky oak
#

yeah, seems to be

dense tapir
#

you have to use those versions I screengrabbed and posted

smoky oak
#

LMFAOOO

swift harbor
#

Any hints on the washed out results I have shown above?

dense tapir
#

LOL

#

Dee's Nuts are so famous sought the world over.

smoky oak
#

they are what controls the color processing after an image is generated

swift harbor
#

I am getting that with any model I have tried.

smoky oak
swift harbor
#

Including those I have included the VAE

smoky oak
#

you can see how big of a difference it makes

#

oh, hmm...

#

@dense tapirWhat was that addon you were using that was trashing contrast?

swift harbor
#

I am putting the VAE in the SAME folder as the model.

dense tapir
#

Dynamic Thresholding

smoky oak
#

A1111 is by default set to None I think

dense tapir
#

I never could get it sharp enough

smoky oak
#

you can enable auto mode

swift harbor
#

Aha! How do I do that?

smoky oak
#

just a sec

dense tapir
smoky oak
#

@swift harborSettings

#

user interface

#

you wanna add the sd_vae

dense tapir
smoky oak
#

and then apply and reload UI

#

yeah, it does what is shown up there ^

swift harbor
#

Ahhhh, I would NEVER have guessed that. Thanks!

smoky oak
#

you can select automatic mode, which I think looks for VAE's that share names with checkpoints

dense tapir
#

It does

swift harbor
#

I have noticed many VAEs have random names that don't match the checkpoint.

smoky oak
#

You can see they make a monumental different

swift harbor
#

Monumental indeed!

dense tapir
#

Correct so I never use auto

smoky oak
#

I use the same VAE for everything

swift harbor
#

Can I rename the VAEs to match the model?

smoky oak
#

I find it just looks good across everything

smoky oak
dense tapir
#

yep

smoky oak
#

I use VAE-ft-mse-840000-ema-pruned, which is from open AI

swift harbor
#

What about LoRas? Can I rename those as well?

smoky oak
#

yessir

#

their trigger words will stay the same

#

you can rename models as well

#

all of my models are renamed for organizational purposes

swift harbor
#

Hey thanks! I was afraid of breaking things. LOL

smoky oak
#

no worries haha

swift harbor
#

Yes, they put version numbers that make it hard to keep track when typing

dense tapir
#

I like to sort them based on V1.x or 2.x

smoky oak
#

just make sure to keep the file extension (.ckpt, safetensors)

swift harbor
#

Of course. 🙂

smoky oak
#

those little reloads. There is one with LoRA's as well

swift harbor
#

I didn't know Loras needed to be reloaded.

smoky oak
#

if you add new ones or rename them. It just refreshes what it sees in the folder is all

dense tapir
swift harbor
#

I just add them to the folder, and Auto1111 just uses them without error.

smoky oak
#

that will work if you add them before it launches

swift harbor
#

I has worked for me even after launched. I was surprised too.

#

As long as I get the name right, Auto1111 just finds them

smoky oak
#

ohhh, do you not trigger by clicking?

swift harbor
#

No

smoky oak
#

the reason we have the layout is cause if you click them, it adds in their trigger word

swift harbor
#

I only click the reload button when I add new checkpoints, not loras

smoky oak
#

if you click them, it adds them in, or removes them

#

if you go into the LoRA UI in SD, you can just click on them to trigger them

swift harbor
#

I didn't know that. LOL

smoky oak
#

Yeah haha

#

its super time saving

dense tapir
smoky oak
#

click them again and it will remove the LoRA

swift harbor
#

I probably don't even know where. I am sick and tired of searching for the files to recall the proper name. I am sure there must be an easier way

smoky oak
swift harbor
#

LOL it was the WHOLE TIME, and I didn't know! LOL

smoky oak
#

yup haha

#

its way easier to use like that haha

#

it keeps all TI's Hypernetworks, ckpts, and LoRAs

swift harbor
#

Thanks goodness, that was driving me CRAAAAZYYYYYYYYYYYY

dense tapir
#

That is the sort of thing I want in comfy to make it actually comfy

smoky oak
#

glad I could help haha

swift harbor
#

Very much, thank you!

smoky oak
#

No worries :p

#

I try to give back to the community as much as possible haha

swift harbor
#

The washed out images were driving me nuts as well

smoky oak
#

did the VAE fix help?

dense tapir
#

These images I am lovin' it.

smoky oak
#

I know a lot of less realistic models struggle hard from no VAE

#

some even crash

swift harbor
#

I don't know yet. I have to see where in the settings, using your screenshot and see.

smoky oak
#

for reference, if you ever use Anything Furry V1 (best generat anthro model), it will crash with no VAE

swift harbor
#

I was lucky so far - no crashes for lack of VAE. In my innoncence, I assumed I could just place the VAE on the same folder, and that would do it, but noooooooooooooooooo

#

They have to be ACTIVATED. No one says anything about that at CivitAI.

#

They just tell us to put it in the same folder.

dense tapir
swift harbor
#

Thanks again for making life easier again. Auto1111 is good, but only if we know these details. 🙂

smoky oak
#

But I have heard of it causing issues with other models

swift harbor
#

I am trying to keep models under control - SSD can't afford such sizes.

#

They are crazy large for SSD standards.

#

Can we change the location of models to another drive, where I have more space?

dense tapir
#

symlink

smoky oak
#

And yet I only use three of them

#

But I always have the other ones in case

#

But really though, I really need to delete them to save the space. I have outrageously fast internet, I can download them in like 10 seconds

dense tapir
#

I have all of my models on the HDD symlinked to the SSD

swift harbor
#

A "just in case" that kills our disks like they were nothing.

#

I am assuming symlink is yet another A1111 extension?

dense tapir
#

no

#

symbolic link

smoky oak
swift harbor
#

Like a shortcut?

smoky oak
#

It links multiple drives together

dense tapir
smoky oak
#

If I remember properly, it might even cache files on the SSD temporarily that are commonly accessed from the hard drive

swift harbor
#

My system drive is only 1TB, so I don't have that luxury.

smoky oak
#

Understandable haha

#

I'm actually going to delete my extra models right now

swift harbor
#

I also should be doing that, since KNOW which ones I use the most. 🙂

#

The problem is the "just in case" LOL

dense tapir
stark vine
#

link to the upscaler?

fiery lichen
stark vine
#

if that's the one that's pretty interesting

#

although doesn't look like something Topaz can already do?

dense tapir
#

No, gigapixel/topaz hasn't updated their models in a long long time now

#

Not sure wtf is wrong with them

stark vine
#

I'll have to check it out

#

anyone knows if it works better for inpainting than lanczos?

dense tapir
#

Oh, my

swift harbor
#

Is there a way to find out to what model a VAE belongs to when they come with random names? X___x;;

dense tapir
#

Nope

swift harbor
#

I knew it

#

Geez

dense tapir
#

Makes it hard to track them down

swift harbor
#

Indeed, pretty much

#

Can a VAE be used between different models? Is there a dependency?

#

I guess I will try and try and find out. 🙂

#

I have 45 models, but only 7 have VAEs.

dense tapir
#

no dependancy

swift harbor
#

Ohh, I see some chance of improving the looks on some models that looked washed out then. ^_____^

dense tapir
#

Yes

swift harbor
#

I was fixing this in postwork! LOL

#

But NO MORE, now that I know how it works.

dense tapir
#

Well, post work can help it even more

swift harbor
#

Well, I still have to fix the hands. LOL

#

And many times the eyes

dense tapir
#

even a photoshop auto tone/colour/contrast for the lazy can

swift harbor
#

I don't trust anything "auto" in my postwork.

#

Sometimes auto-contrast works, sometimes it destroys the image. Better do it by hand. It's faster. .

#

Auto settings are like barcodes - they are great when they work. LOL

dense tapir
#

just depends on the mood

swift harbor
#

Too much post

#

At least in my taste. 🙂

dense tapir
#

I like the look of the too much post over the blah of #2

swift harbor
#

LOL

#

I have this VAE called "vae-ft-mse-840000". WTF is that supposed to be?? LOL

dense tapir
#

For me I am not going to spend hours refining something that I am not getting paid to do. Just not my thing as I can spend the time better elsewhere.

#

that is the master one

swift harbor
#

What is a "master VAE"?

#

The ring that rules them all?

dense tapir
#

pretty much

swift harbor
#

Ahhh good to know!

#

I will rename it to MasterVAE. Makes sense to me.

dense tapir
swift harbor
#

Thank you again for the help. 🙂

dense tapir
#

You're welcome

swift harbor
#

Is a LoRa version of a full model comparable?

#

I mean, 5GB to 200MB. What can we expect from that?

dense tapir
#

you can expect quite a lot actually.

#

a lora isn't really a full head on model though

#

DB>Lora>HN>TI

#

I squarely blame SAI for the death of hypernetworks for never having refined them.

#

they were so good but took so long to train as they had no optimizations

#

I still, to this day, use my HN I trained as a general all around. If I could find the data I would see to train them as lora

swift harbor
#

So, are you saying the ARE comparable, even if LORAs are not full models? What gets lost in the way? I guess that's what I was wondering.

dense tapir
#

Nothing gets lost

#

Lora is just a different training method

swift harbor
#

Sounds like magic 5G -> 200M

dense tapir
#

LOL, no

#

A prime example is you ask for Spuds McKenzie. Does the model give you something that looks like him but not quite? A body double? Cool, train in Lora and now it knows it precisely. No? Ahhhh, well DreamBooth full train a new model.

swift harbor
#

Some of these LORAs are indeed a hit & miss

#

I only know after I try them

dense tapir
#

I have no idea about all the shit out there (and there is a ton of it) for 1.x as I find much better out there for 2.x

#

The day 2.0 dropped I tried it and dropped everything from 1.x

swift harbor
#

Some LORAs are made with 4 images only. Some with 800. I wonder what is the sweet spot?

dense tapir
#

that which works. I am being serious.

swift harbor
#

So it's not about quantity, but quality?

dense tapir
#

BINGO

swift harbor
#

Makes sense.

#

Probably images that show the most different sides of the subject?

dense tapir
#

I do styles though and it takes a lot more than on a subject

#

Styles sweet spot is 50-300

swift harbor
#

How long on a RTX 3090?

dense tapir
#

10-15m

swift harbor
#

Let's say, 300 images

dense tapir
#

10-15m

swift harbor
#

Impressive.

dense tapir
#

Friend trains on his all the time

swift harbor
#

What's the image size for that?

dense tapir
#

768 sq

swift harbor
#

So you have to crop them all manually?

dense tapir
#

We only do 2.1

swift harbor
#

That's high res, so it makes sense it has to be 768

dense tapir
#

We do as we both do not trust the invisible machine to do it called buckets but others use buckets

swift harbor
#

I only know render buckets, don't know what it means for AI training

dense tapir
#

buckets are nice if you have a lot of weird shaped images and I mean a lot of them

swift harbor
#

Like fractal images?

dense tapir
#

each bucket is a resized image it did for you to match your dimensions in latent space

iron horizon
#

Hi, I just joined the server
Someone can explain me where can I generate images?

dense tapir
#

Not on this server

iron horizon
#

Oh crap

#

Where then?

dense tapir
#

We all do it locally

iron horizon
#

How does that work?

#

I'm pretty new in these things

swift harbor
#

One more victim of this timesink. Say goodbye to your friends and family. LOL

dense tapir
#

HAHAHAHAHA

swift harbor
#

And buy a bigger SSD!

dense tapir
#

See, I honestly have little enjoyment from generating (inference) and a whole lot in training but colab dang near killed me. I was so burnt out

#

My 1060 can't do it

swift harbor
#

Wow, I didn't know those even had Tensor Cores

dense tapir
#

LOL, they don't on a T4

#

10m 3090 same exact data on T4 is 25-30

#

plus 10 more minutes just to fire it up

swift harbor
#

I can only imagine. Never trained my own model... yet.

dense tapir
#

I did DB in 1.5 and tons of TI embeddings on 2.x that I released and having a lot of trouble with Lora due to 2.1

#

2.1 fights ya

swift harbor
#

I have a gallery with 17 years of my 3D renders. I was thinking to making a model of my own stuff.

dense tapir
#

See, just to show you how I am, as opposed to @smoky oak who does models I saw 17 years of 3d renders and drooled to get them and make them a style.

swift harbor
#

That was exactly what I was thinking of - making a style of my own work.

#

No idea how, though.

dense tapir
#

The issue with styling is finding a common theme among them.

swift harbor
#

Like what?

#

What would a theme be?

dense tapir
#

You need an eye

swift harbor
#

I have 2!

dense tapir
#

You can't throw picaso and Banksy at it as it would come up with shit or something inbetween

#

most times shit

#

no, an eye for see the common

swift harbor
#

I did mostly pinups, so that sounds like a common theme.

dense tapir
#

like how critics can see a painting and tell you who made it

swift harbor
#

People tell me they recognize my 3D renders, so I likely have a style.

dense tapir
#

Well, show me three random ones in private if you wish and I will see what lora might do with them

#

need at least 20 though

swift harbor
#

I have 2000

dense tapir
#

if you have 100 all the same theme/style you are golden

swift harbor
#

17 years is a long time to render. LOL

#

I certainly do.

dense tapir
#

well, do you think you have 350?

#

Notice I asked 350

#

350 all the same style/theme

swift harbor
#

I think so.

dense tapir
#

you probably do

swift harbor
#

Even more in the same style.

dense tapir
#

No more it makes Lora vomit

swift harbor
#

LOL

#

So 350 is a good number for style?

dense tapir
#

chill

swift harbor
#

Still have to crop them all manually? O_o;;;;

#

350 is a lot to process

dense tapir
#

Where you will find a pita with is the tagging you must do

swift harbor
#

Thanks, I am keeping this link for later research. ^^

#

Maybe I can automate the tagging with Python?

dense tapir
#

No, it is never as good as you are

#

One day I hope

#

It loves say 1girl yet there is 1boy

#

or vice versa

swift harbor
#

There are virtually no men in my renders.

dense tapir
#

That was an example

#

and to it you see girl it sees man

swift harbor
#

I guess I don't like men as much as I like WOMEN. LOL

dense tapir
#

iow you have to fix it

#

Now that second link read it first and play with it

swift harbor
#

Oh, so SD interrogates the images, tags them with what it THINKS is there, and we have to fix them by hand?

dense tapir
#

yeah, via an extension

swift harbor
#

Now I understand why all these LORAS made with 4-10 images. LOL

dense tapir
#

some people have no loved ones so they sit in their basements for months on end and just do it by hand. They have become the machine.

swift harbor
#

One day archeologists will find these mysterious mummies on basements, and wonder what they were doing there.

dense tapir
#

I can hand them a picture and in seconds they will have a tag. I am like wtf?!?

#

LOL

swift harbor
#

And I thought cropping was the hard part!

dense tapir
#

full finetuning they spend months at a time on doing 300k images and tagging

swift harbor
#

Not for humans....

dense tapir
#

not for me at least but they do it

swift harbor
#

One day, AI should do that part - but then we will be useless anyway.

dense tapir
#

A micro finetune is consider 1k, or less, images and captions

swift harbor
#

That's crazy

dense tapir
#

yep

swift harbor
#

I would want to automate that with Python.

dense tapir
#

75k is the average for finetuning

#

but you can't, not yet

swift harbor
#

Yes, I can see that now

dense tapir
#

the tech is getting better but by hand it is just far better

swift harbor
#

Based on my experience with image interrogation, I can see it's far from good enough

dense tapir
#

When the day comes we slide an image to the ai and it says Jennifer Anniston in blah blah standing next to blah blah we will have finally arrived

#

right now it gets a lot right then so much it doesn't

swift harbor
#

That will probably be "Jennifer Anniston mummy" by then

dense tapir
#

six months ago it was all rubbish

swift harbor
#

I am impressed how fast AI progresses.

#

What used to take years to advance now takes months

#

Sometimes weeks

#

But I am happy enough that AI still won't replace me.

dense tapir
#

Evetually nothing is safewhen AI/robots can already replace some surgeons.

smoky oak
#

It's an extension of which adds a script that you run an image to image mode

swift harbor
#

Ai can already replace lawyers - well deserved!

dense tapir
#

I do not trust AI until it can learn on its own and purge any bias it was programmed with from its devs.

smoky oak
#

I've done 3D renders/art for around 5 years now as well

swift harbor
#

I am the one who has done it for 17 years. 🙂

dense tapir
#

Why in the fuck would you be confused? You train models I train styles. I see it I don't care about the models I want those styles you probably want the models.

swift harbor
#

One question - can I put VAEs into a subfolder?

dense tapir
#

There must be a memleak as the more I gen I will eventually run out of vram and have to close it down and start over

swift harbor
#

Would A1111 still find them?

dense tapir
#

they are already there

swift harbor
#

I have put mine in the same folder as the models.

#

They tell us to do it that way.

dense tapir
fiery lichen
swift harbor
#

I know we can put LORAs in subfolders, but what about VAEs?

dense tapir
#

Did you see?

swift harbor
#

I don't know who's talking to whom anymore. LOL

dense tapir
#

Those images were to you

#

that is how mine is set up

#

those are all on E a HDD and Auto is on D the SSD

swift harbor
#

I don't get it - what does the upscale extension has to do with this?

dense tapir
#

It doesn't that is another convo

smoky oak
swift harbor
#

LOL

smoky oak
#

You install it from the extensions tab

swift harbor
#

So... we can place the VAEs on a subfolder?

#

I will restart A1111 and find out. 🙂

#

If I put VAEs on a subfolder, A1111 can only "see" half of them for some reason

#

I wonder why?

smoky oak
#

Hmmm

#

No idea

#

That's weird

swift harbor
#

Weird indeed

#

I have 7, it sees only 3

fiery lichen
# smoky oak You install it from the extensions tab

it is not on the collabs i am using. I guess I have to install that one and try it out. I can install extensions when i have the github link. But as far as I can check, it did not come with my collab as standard.

swift harbor
#

Ah, I see - it only sees the ones that end with "xxxxx.vae.pt". It won't see the ones named "xxxxxxxxvae.safetensors"

#

It only sees them if they are in the same folder for some reason

#

Oh, I got it now. I have to rename them to "xxxxxxxxxxxxxxxxx.vae.safetensors" The dot is important.

#

Ok, now it sees them all. Strange that it could see them as they were when placed in the same folder.

#

But require renaming if placed in a subfolder.

#

Great, now I can separate my VAEs from the actual models.

dense tapir
swift harbor
#

People told me the generic VAE for realistic style is the 84000, while the Kenshi VAE is generic for Anime style.

dense tapir
#

Kenshi is anime

#

can try vice versa if you like

swift harbor
#

I thought VAE was color correction. I didn't think the styles mattered

dense tapir
#

You are massively overthinking it. Anime has their own colour scheme and flatter

swift harbor
#

But I assume it does matter because Anime style has a brighter, more vibrant color scheme

dense tapir
#

that as well, but not always

swift harbor
#

Yes, Anime style varies a lot

#

There is no ONE Anime style. There are many

dense tapir
#

I prefer dark themes and highly saturated black

#

I get too many colours I bail

swift harbor
#

So now I know what these VAEs will do. The Kenshi one will give more vibrant and possibly more saturated colors

dense tapir
#

Could be the other way too

swift harbor
#

I am new to VAEs, so I will have to play with it to see. 😉

dense tapir
#

We don't use vae in 2.1 as they are baked in. Sad about that really.

swift harbor
#

Until now I was fixing colors in postwork, since I have to fix hands anyway.

dense tapir
#

SDXL I have no hopes for

#

I don't even care for the name as it sounds like a used car dealer trying to unload a 1977 Ford Pinto for 20 grand on me.

swift harbor
#

I am rarely using 2.1 models so far.

#

I am not even sure which model is which

dense tapir
#

I prefer 2.1 for inference and for training 1.5 due to the CLIP. CLIP is fantastic, and can actually count, but fights you for training

swift harbor
#

CLIP is the image interrogation part? Or the text to image?

dense tapir
#

TE

#

text encoder

swift harbor
#

text to image?

dense tapir
#

different

#

Hard to explain but TE sucks ass juice in 2.1

swift harbor
#

I used to know what CLIP means, but my brain decided there are more important things to place there.

dense tapir
#

Yeah, OOM error so dump this to make room

#

no upgrades yet but Elon is working on that

swift harbor
#

Pretty much. Like my SSD

#

Space is limited, so use wisely

dense tapir
somber sorrel
#

LoRA name?

swift harbor
#

If I wanted to make a model hash search Python app, is there a repository to access models by hash code?

dense tapir
#

Not that I know of

swift harbor
#

I know CivitAI can do it, but I wonder where they are getting the info from?

dense tapir
#

Not sure I like the idea of a central repository for said hashes

swift harbor
#

Sometimes prompts only provide the hash code, and I want to know what model that is.

#

I usually go to CivitAI and search there

#

Would be nice if I could search directly from a Python script.

somber sorrel
swift harbor
twilit marlin
#

Can anyone tell me how to embed hugging face embed onto stable diffusion webui google colab?

swift harbor
#

Have you tried PNG Info?

#

If it's something that came straight out of A1111, PNG Info will extract the prompt for you.

twilit marlin
#

im so confused I ran the command and its not in the embed list

swift harbor
#

I did this long ago when I first installed A1111, and I can't recall how it works. But I do recall they give instructions about this.

twilit marlin
swift harbor
#

You mean the code they give you to use their model?

dense tapir
swift harbor
#

It works like an activation key that recognizes you as the user.

swift harbor
# dense tapir

LOL I tried making characters ride bikes before, with hilarious results.

twilit marlin
#

Im trying to embed one of their models from hugging face

swift harbor
#

Embed? Isn't it just placing it into a specific folder?

twilit marlin
#

do I download the .pt and put it into the folder then run the command?

swift harbor
#

If A1111 is already running, you first need to press the refresh button to make the program see the new file.

ripe nimbus
#

how could i

dense tapir
ripe nimbus
#

use the stable thigs

twilit marlin
#

so all I gotta do is put it in the folder?

swift harbor
dense tapir
swift harbor
swift harbor
#

One day.... One day....

dense tapir
#

Well, controlnet allows good hands

#

I don't use controlnet though

swift harbor
#

Welcome to the world of mutants!

dense tapir
#

Funny thing is I make a mutant it has human hands I make a human it has mutant hands

swift harbor
#

For those lunatics who claim AI images are "easy" and take "no effort", they obviously haven't tried it yet.

#

Hands are getting better with some helper embedding now. Better than nothing

#

I refuse going to Blender to pose a hand, extract the skeleton pose, and bring it back to A1111 hoping the perspective was right.

#

Not for everyone

dense tapir
#

I hate blender for decades now. No thanks to that.

swift harbor
#

I started with 3DSMAX, so Blender feels like torture to me. Can't help it.

dense tapir
#

Cinema4D and pre Autodesk Maya were my favs

#

3dsmax would be my 3rd

swift harbor
#

What I can do in MAX with 10 clicks, it's 100 in Blender.

dense tapir
#

Blender was never really made for what people use it for though.

#

there is a reason its UI sucks balls

#

People seem to love it though

swift harbor
#

The UI got better, but the workflow sucks if you have ever used other 3D programs. Very inefficient to me.

dense tapir
#

That as well

swift harbor
#

People who love Blender have probable never used anything else. You can't miss what you have never had,

dense tapir
#

C4D, or 3ds or pre AD Maya was so simple to get around in

swift harbor
#

Comparing to Blender! LOL

dense tapir
#

Why this driving need to make shit overly complicated?

swift harbor
#

Same thing I asked myself. Good to know I was not alone on this.

#

Ken in Blender: I want to edit this vertex. I can't!!! Oh, I have to switch to EDIT MODE....

dense tapir
#

Hell, I see it in Windows where something you used to just right click to get to the next version is two then the one after is so damn deep you can speend a lunch break getting to it

swift harbor
#

The art of Windows "improvement".

dense tapir
#

Yeah, lol

#

wasn't like that until ballsack took over as ceo

swift harbor
#

It's like arranged marriages - you will love her over time.

dense tapir
#

some small changes but then he came in and it has never gotten better

#

11 is so deep on shit as if they are trying to hide stuff on you

swift harbor
#

I can't use it. Haswell-E here.

twilit marlin
dense tapir
#

Nor I but I see others and vids and just wow

swift harbor
twilit marlin
#

.pt are embedding files right?

#

Ive put it in the folder and restarted

swift harbor
twilit marlin
swift harbor
#

The PT extension means nothing.

twilit marlin
swift harbor
#

Yes

#

But remove it from the models folder too. 🙂

#

If it's still there

twilit marlin
#

yea it ist

dense tapir
twilit marlin
#

isnt

swift harbor
#

Then you should be fine

#

BTW, you don't have to restart A1111 for this

#

If the file is there, and you use it on a prompt, A1111 will find it.

#

Same for LoRAs.

#

Just models require pressing the refresh button

dense tapir
swift harbor
#

Was looking at some prompts from CivitAI, and some images have sizes like 600x512. HOW????

#

That size is not allowed....

wintry nebula
#

image

swift harbor
#

I can't reproduce these images because A1111 won't allow 600 as width

#

It has to be multiples of power of 2

#

The VAE files make a world of difference. This looks edible.

#

No mutant fries! LOL

#

Had to try several different VAEs, since the 84000 and Kenshi looked both too saturated to be real.

wispy nest
swift harbor
#

I tried, but A1111 auto changes them to the closest power of 2

#

Ah, I have to hit ENTER! Doh!

wispy nest
swift harbor
#

Yeah, I didn't do that.

dense tapir
swift harbor
#

Some prompts include the word "BREAK" instead of periods. Does that make it any different?

dense tapir
#

does matter

#

breaks being UK and periods American

#

this was trained in America so a lot of those words it has no idea

#

man, is there no batch interrogator yet in auto?

#

have a nice one for tagging

swift harbor
#

Funny that I can make the AI make me a burger with fries, but if I ask it to put ketchup on the fries, it sticks the fries inside the burger and throws it on me. Sorry for asking.... <___<;;;;

#

On my 10th try, it put the ketchup in the soda.

dense tapir
#

I see one so let's give this new extension a try

#

the thought of that makes me sick

#

milk in pepsi was a think but ketchup in coke

swift harbor
#

Point taken, the models have no clue on how to put ketchup on fries. But it can put it on the soda cup instead.

#

If the AI has never seen it, it probably can't do it.

#

First time I put ketchup on fries in Brazil, people almost threw up when they saw it. They told me I ruined it. Well, to each, their own.

#

Is "bad-artist" an embedding? It seems used a lot.

#

Oh well, it IS an embedding. Better get it

dense tapir
#

I don't use 1.5 so unknown to me

#

Pretty cool I see an extension to change my ckpts to safetensors which I was wanting.

swift harbor
#

Darn, there is another one: "bad-image-v2-39000"

#

Where do I find that?

#

CivitAI and HuggingFace don't know it.

livid ermine
#

hey, what y'all think of my two girls? one norwegian and one malay

#

and how realistic are they compared to the most realistic ai photo you've ever seen?

swift harbor
#

They look good, but the eyes give the blonde away, and the fingers on the Asian.

livid ermine
#

interesting, thanks man!

swift harbor
#

Those are the usual suspects with AI. Hard to get it right, mostly luck sometimes. 🙂

#

Sometimes I get perfect hands, but the face is all wrong. Darn it! LOL

livid ermine
#

could try again with GFPGAN regarding the eyes, and i saw someone mentioning hands, what was it again

swift harbor
#

It's very hard to fool a human when it comes to human faces. It's engrained into the DNA how to detect the smallest flaw.

livid ermine
#

oh yeah openpose hand

swift harbor
#

ControlNet

#

You can also try "bad-hands-5" embedding to improve hands, but it's a hit or miss.

livid ermine
#

hmm

swift harbor
#

AI seems to like spaghetti a lot when it makes hands.

livid ermine
swift harbor
#

Dreambooth has a lot of limitations

livid ermine
#

i see.. in what way?

swift harbor
#

I haven't seen it in a long time because I have A1111 installed locally. But last I've seen it, most options were hidden

#

we could only change very basic things

livid ermine
swift harbor
#

Control net is the most powerful extension I have ever seen for A1111.

livid ermine
#

i see

#

oh really? thats good!

swift harbor
#

Tons of vids on YouTube, with people drooling all over it. Very worth your time.

#

It allows things that were not possible in A1111 before.

#

Like this with a single prompt.

livid ermine
#

definitely checking it out

#

i wonder where to put that stuff in my script though..

swift harbor
#

It has a part just for hands, integrated with Blender 3D.

#

Extensions.

livid ermine
#

so i just add that stuff on top like gfpgan? and not integrate it with the dreambooth training somehow?

#

glad i could use ruby for this though. pretty isnt it? 😄

#

and i have a little secret too

#

man those are some nice pluggos imho (my girls above had double exposure to film grain)

#

or are there better choices available here in the AI world?

#

its like with my music, my plan is always to add so much vinyl emulation and analog distortion so that you can't tell it's digital and synthetic

swift harbor
#

They have tons of models to choose from at CivitAI. More than I can keep track of.

livid ermine
#

cool ill check it out, cheers!

#

burzum, the famous black metal guy is who that scripts for. my godfather shared jailcell with him once...

swift harbor
#

Any hints on where I could find the "bad-image-v2-39000" embedding?

swift harbor
#

Thanks!

#

When I go to Hugging Face, and search for "bad-image", I get zero hits

#

I must be doing something wrong

livid ermine
#

theres so much how to make old photos new but not the other way around afaik

swift harbor
#

There might be. Sometimes the challenge is finding things.

livid ermine
#

🤣 yeah true

#

so true

swift harbor
#

Good example right above - I have searched for this embedding at Hugging Face, but didn't find it, even though it was there.

livid ermine
#

aka. the art of browsing, very important concept in ecommerce psychology i learned the other day

#

music too, listeners are way more likely to share if they stumbled upon it themselves

#

what are you trying to achieve ken1177?

dense tapir
swift harbor
#

I am trying to replicate the sample prompts from the HELLMix model. But I am missing some embeddings.

livid ermine
#

cool

swift harbor
#

Finding missing embeddings has proven to be a challenge. I still don't know how to find that last one if I had to search it myself. I went to "models" and searched, but not found

livid ermine
swift harbor
#

This one was "experimental", so it might be hidden somewhere

livid ermine
#

sorry thats not helpful

swift harbor
#

LOL

#

There was a street in Ohio called "Manko", which means "vagina" in Japanese. LOL

livid ermine
#

haha

swift harbor
#

I took a picture, or my friends wouldn't believe it

#

Now back to the embeddings from the negative prompt, they can change the final image completely.

livid ermine
#

vikings liked to give funny names to islands around scotland etc 😄