#🌶|off-topic
1 messages · Page 9 of 1
Well yeah, part of the AI art community makes it worse than it "should"be
The point isnt to ban AI anyway
Nobody wants that
With exceptions
Can anyone recommend software to upscale videos locally? I have some old video from a 2003ish digital camera that i'd like to see if I can improve
They could of just stole the real things and leaked them. At least that didnt happen.
Hell they could of prob hit it and put a ring on it but not now. dummies
im pretty new sorry friend
i post my art to instagram ill follow u if u got one
it used to be photoshop, nothing new
the amount of knowledge it takes to generate decent looking deepfakes exceeds the amount of knowledge it takes to make decent fakes in photoshop,.. IMO
totally unrelated just something ive done in blender
a rotating mobius strip looks really trippy
it doesnt look like its rotating to the point i had to add a wireframe shader
they dont even look like theyre rotating at the same speed (they are i checked)
my eyes...too bright
ah shoot. I was going for like a clean low contrast look.
I'll make it more dramatic or something if I rerender
id love to try the stable diffusion blender addon on it. idk how to use it
Nah
I mean the nearest would be one with technical skills ofc, but with Photoshop you would need to be good in several areas + its own tools to make convincing deepfakes
learning to run and configure a stable diffusion server to do that with seems harder than slapping a face on a naked body in gimp but maybe thats just me
lets start a youtube channel that is like a parody of flat earther type stuff and also the anti flat earthers like a fake skeptic channel
but our gimmick is 'were the real skeptics, cuz we'll believe ANYTHING, the other skeptics dont believe anything and theyre fake'
where the joke is that we just uncritically believe any bullshit but call ourselves real skeptics
chatgpt is alive? i believe it
chatgpt is people pretending to be ai? i believe it
needing sleep is a made up concept to waste our time
slapping a face on a naked body in GIMP is far from a deepfake, it sounds more like "Photoshop level over 9000" memes
so it's easier for a highschool or to learn to make deep fakes than create believable photoshops? I'm just trying to figure that out. I haven't been a teen in two decades
For Photoshop you will need to know the tools and techniques
You kinda need that for AI, at elast for now if you want to make it look decent
Depending on the tool, yes
listening to NPR right now talking about the deepfakes
tech news today episode 310
around the 50 minute mark
"it's all fake, it's your face but the rest is fake, so clearly we need to limit AI capabilities"
https://www.gamesradar.com/ai-game-coding-tools-instantly-result-in-angry-birds-clone-and-opens-some-potentially-dangerous-floodgates-for-mobile-storefronts/
I hate how negatively contotated this headline is
as if Angy Bird clones didn't exist without ai, or that AI inherently is dangerous but non AI isn't
Sometimes stable diffusion is slow af!! and sometimes its lightining fast. What am i missing here?
Big tech races to dominate the next era of immersive computing. Is @Meta's bet on smart glasses & the #metaverse enough to beat @Apple? Meanwhile, how will @starcitizen innovations impact the open metaverse?
#VR #AR #spatialcomputing
https://youtu.be/yrhrTFazq7A?si=PSWX-_xeOIkFN12d
Description: Meta is pivoting to the metaverse with new VR headsets and smart glasses, while game studio Star Citizen innovates with a massive multiplayer engine, raising questions about who will dominate the next era of immersive computing. The article explores how Meta's focus on mixed reality and AI aims to make smart glasses a computing plat...
i had to look that up. it's not an analog chip so i dont know why they call it that. also, it's specialized ONLY towards image classification purposes. So i guess the screen shotted text is claiming 3000x faster there, which the nature paper doesn't reflect
smart glasses have been on the market for years and nobody uses them. the interfaces suck. the screens in your glasses are not nearly as high fidelity as your phone. the form factor is not something that's usable. you also can't show people what you're seeing on your glasses as easily. it's gimmicky af.
apple's vision is cool but microsoft has already been shaping the spacial computing paradigm for years already. the hololens is a real product and microsoft sells every single one they make. It's just, industry like automotive or military buy them all up. Spacial computing is already here if you're in the right fields. And you can get a hololens today if you fork out the $5000 for one (might be more)
Anyone have a spreadsheet/a recent tier list of GPU's to make an LLM server, looking to do it on budget in the future expanding. Max budget rn being $2000~
Is loincloth NSFW?
My computer make a sound like it is some hardware is found or a USB disk been activated, but I do not get any notification for what make that sound, so do anyone know how I can check up what is activating that sound?
The same happens to me
What I hate about the art community is that you have to be artist at all. I'm not an picture artist, more like programmer/worldbuilder (and trying to be a writer), so my art is really just secondary. The reason I use dress-up/ai-art is because I want to focus on what matters for me, which is worldbuilding and writing.
Wdym with the relation to art community? That they attack you for not being an artist eventually?
I think btw artists should give programming a chance because it brings artists to another level because of what they can achieve with scripting for example
Used to, more precisely. Back when I was more into Gacha Life. Yeah, I know it's not a substitute for a real drawing, but that was never the point.
imo its a complicated situation
i definitelly got to know some through and through anti AI hardliners
i dont like their stance although to a certain degree im on their page
you cant satisfy those tho and the energy waste isnt worth it imo
Hey! See what AI can do in gaming! We've AI for visuals and animations. I will be glad to hear your feedback.
[iOS: https://apps.apple.com/us/app/cat-journey/id6471379757]
[Android: https://play.google.com/store/apps/details?id=com.yaskravicontent.catjourney]
https://youtu.be/tTceryLuPVE?si=PxvdEff2q_u3w0J3
download here:
Android: https://play.google.com/store/apps/details?id=com.yaskravicontent.catjourney&hl=uk
IOS: https://apps.apple.com/us/app/cat-journey/id6471379757
Dive into a captivating world with the red cat in the spectacular 2D action-platformer "Cat Journey"!
📜 Enthralling Story: The fate of the Cat's sister is in peril! Join him on a...
Hey! Help us do our job for free! It's free!
heyoo, did I just see someone at devconnect in Istanbul with a stability ai shirt 👀 ping me in case you see this
did auto11111 also feature voice ai?
or it was the another one
developer
Happy day. Strong project and friendly community.
Good luck to all 💸 💰
Been looking at weird AI models and this one caught my eye pretty quick
The clips on the right are 3d animations generated by our AI.
Get access to the AI software here: https://concertcreator.ai
Turn any audio file into 3d virtual piano concert with one click.
The AI extrated notes from raw audio and generated the appropriate playing technique, hand and body animation.
Clips used in the...
Rather unusual but very interesting software
fascinating
However, as I look further into it, it shut down a year ago in july.
Thanksgiving on Mars, created with Dolle3, Pika Labs, Chirp, Stable Diffusion Audio https://youtu.be/anEuCM0jZVs?si=Fm9LZ22osQ6zDZ9v
🚀🌌 Join us on a captivating journey to the Red Planet in "Thanksgiving on Mars: Echoes in the Tunnels," a thought-provoking sci-fi tale that intertwines the future of space exploration with the timeless themes of cultural exchange and historical reflection.
In this gripping narrative, follow Dr. Wihakayda Chepi, an anthropologist, and SAGA, a s...
stability ai should fuse with openai now that they're vulnerable
Epic
id love to model an articulated recreated of. battle bot thingy
like "rigged" idk how any of that works though. id love a Kerbal space program "like" battlebots video game tho
That would be cool 👀
how can i load this checkpoint in sdxl by Python? https://civitai.com/models/188525/pixar-xl?modelVersionId=211735
Here is my code:
import torch
from diffusers import StableDiffusionXLPipeline
# Passe den Pfad zum Checkpoint an
checkpoint_path = "/home/yago/stablexl/PixarXL.safetensors"
pipe = StableDiffusionXLPipeline.from_pretrained(
checkpoint_path, # Hier den korrekten Variablennamen verwenden
torch_dtype=torch.float16,
variant="fp16",
use_safetensors=True
).to("cuda")
prompt = "a photo of an astronaut riding a horse on mars"
image = pipe(prompt).images[0]
# Speichere das generierte Bild im Ordner "calender"
image.save("calender/generated_image.png")```
could you perhaps post what error you're getting?
Can I install stable Diffusion 1.5 on Intel Core i5 generation with intel irisxe graphics?
Try StableDiffusionXLPipeline.from_single_file instead of from_pretrained
is it typical for prices to increase like this?
this is for an Aorus eGPU
I'd have assumed tech gets gradually cheaper
Are you looking at ru prices? 
Ya, but if it's about sanctions, I'd imagine it've gone way up at the start of sanctions,
ru prices are all over the place, look for global...
there are just too many things affecting ru prices...inflation , sanctions, taxes, oil prices going up... idk, anything can happen there
so...if we're talking about spike - I'd bet on something big, might be global happening, maybe on Aorus themselves.
But...later on it did come back anyway, didn't it?
It looks like it increased slightly over the course of year, but that's completely normal, I'd say even less then I'd expect based on inflation and $ to RUB exchange rate changes
You are buying initial thing in $ afterall , it's components too
Built a job board for AI/ML !!!
Please support on PH
https://www.aimljobs.co/
https://www.producthunt.com/posts/ai-ml-jobs-2
The Best Artificial Intelligence, Machine learning, and Data Science Jobs Only. Best AI/ML Research Jobs Online.
Ya, maybe sense. So maybe ideally wait for a better exchange rate to buy expensive foreign items
inflation, gas \ oil prices affects it too.
(I mean not as an export resource, but just as something trucks needs to buy locally in order to deliver things)
Unless you wait for manufacturers to drop their prices down - it might increase even more by that time.
Darn. Maybe I should just buy. It's gonna be super expensive though, like $1000-2000 depending how much vram I want
+30% import tax from government doesn't make it cheaper too 😄
Ru gov did alot of shit, both for countries around and their own people, so here we are
Foreign things have always been expensive here
Import taxes suck ass, especially in products that aren't made locally
Like, I'd totally buy local if it were an option.
Does anyone know if there's any software/AI tooling that can seamlessly blend from one song to another? Not like a regular crossfade, but like something analogous to img2img or flowframes, if that makes sense, so it's like the song transforms slowly into the other one?
I need to make a monster from my friend's original photo. Can I find a guide somewhere?
just search something about "img to img stable diffusion" or something like that...there should be plenty of guides, it's not that complex
invoke AI made Head Line New! https://youtu.be/d5Kd0Vv_Ark?feature=shared
invokeAI redistribute open source Not safe for work content Which anyone can get their hands on.
But I'm not allowed to post a link When I'm trying to help out
Is it just me or do the SD1.5 models still look way better than the SDXL models?
Did OpenAI leadership fall victim to hubris? Have they jeopardized humanity by unleashing unguided AI models? Is this the worst case scenario? 😱 https://www.youtube.com/watch?v=jclJUJBvFzY
This video covers the shocking firing of OpenAI CEO Sam Altman just a week after their successful Dev Day conference unveiled major AI advances like GPT-4 Turbo. I discuss the philosophical war within OpenAI between the "effective accelerationists" like Altman pushing rapid progress, and the "doomers" worried about safety. Speculation is that th...
no. This is clickbait.
I havent tested sdxl yet (because I only have 3gb of vram 😁 ) but I can generate up to 1024x1024 on sd1.5 and if I use the correct prompts the results are very good, and in some cases better than the ones that I´ve seen on some sdxl images
sd1.5 can go to 4k no problem. There's really no resolution limit. And ya, SDXL always looks blurry and grainy compared to sd1.5
https://www.pcgamer.com/us-gov-fires-a-warning-shot-at-nvidia-we-cannot-let-china-get-these-chips-if-you-redesign-a-chip-that-enables-them-to-do-ai-im-going-to-control-it-the-very-next-day/
this makes me angry humanity should be cooperating to advance technology, not restricting each other
What's their justification?
insecurity paranoia nothing rational
itll just force them to manufacture their own, and naturally outpace the usa. which is funny
more realistically it's just pushing nv card sales underground
and i don't mean that in the future tense, that's already happened
lotta under the table drug gpu dealing
Let's push all trade underground. That way there's less support for criminal activity
making things illegal would 'never' result in a black market being created
RGB image to CMYK pattern in blender 😩
doesnt even control dotsize based on brightness unfortunately
used a mix of stockphotos, stable diffusion and other random images to create this. https://www.youtube.com/watch?v=eRlNxQAFdo4
I exported this video countless times at this point with Universal getting pissy for an extra second of movie, which i had to spend more time re-editing every scene with audio and film. Also dealing with actually errors from premiere pro exporting. well I can't keep rendering it out at this point. I see 3 minor visual an audio errors and I Have ...
But regardless most criminal activity is committed by the U.S. government. If everything goes black market, it means no taxes to fund their wars.
for sure most black market cards will just be arranged by people at the top not regular peopple
GPT2 Says: Jimmy McGee's new YouTube video , "Tall, skinny, and fast : The story of the man who invented a new kind of shoe," is here! The video features many quotes from other people 's opinions about Jimmy 's shoe, but also reveals some personal thoughts about the man himself, especially his own shoes.
No idea why ...
and most criticism of AI boils down to "big companies are misusing it"
hboberguy new plagirism video basically refers to generative ai as plagirism at the end eyeroll
Big companies misuse everything. The problem is capitalism and government, not technology or the freedom to use it.
thats exactly the comment i just left on the video lol
well almost exactly
massive company steals all the art in the world, and then compresses it into a "private" dataset.
if this company simply let the dataset remain open (like the company name) it would at least be less of a problem
the part about policing is good though
hes not entirely wrong throught the video
Laptop 
85c O.O
plz adjust your usage
look into a gpu tweak or something and see if u can limit the voltage or make sure the fan is ramping up
to prolong the life of your hardware
they made a combat robot out of hot dogs
Anyone who knows how to train a U-Net and what might cause it to output BW images?
(despite being given and trained on RGB images)
alright might have been just because i gave it basically deepfried pictures with no detail
Anybody else think reddit isn't having server issues today and the site issues are new systems they're rolling out to prevent site wide scraping?
is this considered vram edging?
no. Shared ram is liekly being used here. You cut that line out suspiciously
shared memory is literally the second graph
although i didnt pay attention to task manager after this screenshot
but training my AI went pretty quickly so i dont think it used shared memory
oh i don't read latineese sorry. forgot thats wher ethey put that.
Okay good job edging that in
its german
its all greek to me
perfection
Hey, what is the most portable high performance (24gb of vram or greater ) desktop or laptop for machine learning that i can bring on a plane that is less than 10 pounds? A mac studio or m3 max? I cant find a laptop with 24gb of vram , max 16gb with rtx 4080( mobile 4090)
i dont know if planes have poweroutlets but i have a ASUS TUF Gaming A17 with a 3050 Ti and 4gb vram and it alone needs a 200W brick
so thats gonna be tough
so a macbook might be your only option if it even has that much vram, i also dont know how fast anything AI related is on the M chip series
My computer has 16gb of unified ram(<=13-14gb of vram) already
apple?
4gb of vram, lol it must be slow like a mac m2 pro
Yes
its not really a burden, just makes me write cleaner code
ah
i was confused of the unified ram
i mean technically i have like 24gb of vram
It means the gpu or the cpu can use most of the ram if it needs to
16gb from GPU and 8 from RAM through system fallback
ah fair
well nvidia can apperently do something similar through system fallback albeit at a strong performance hit though i havent noticed any hit on performance when training rather small models in 32 bit
My safari tabs and os use more than 6 gb of ram but u have onboard memory for the cpu in addition to the vram.
with ai models i mean custom u-nets in pytorch btw
thats nothing compared to windows
though to be fair...
Lol i had code with chatgpt and bard’s help; it used 42 gb of ram, it caused my jupyter to crash
I had a custom UNET called ARM (ArtifactRemovalModel) which used all of my VRAM using only a handful of training and verification images but now I managed to get a smaller and more efficient architecture i like to call MEUNET (Memory Efficient UNET) which is only a fraction of the size (8gb vs now 500mb) on the gpu and only 2mb compared to 120mb yet it outperforms my old model in any way possible, though i have an issue of overmatching and/or unhelpful training data
Oh nice
How do u to optimize memory usage ?quantization and reducing precision and low rank reduction
im training that model at 32bit on my 16gb gpu
its using like 15gb
but for some reason its much more space efficient on my 4gb card
with the same settings
on my laptop its only using about 2gb
Do u also use convolution to compress? How do u denoise then? Same as unet?
yea i use convolution
i mean its a unet
going from 3 channels to 32, then scale down to 128 channels and back up to 32 and then 3 again
designed for RGB in and output btw
specifically trained for removing compression artifacts
I need to learn more code, but i have the ideas
this is a small screenshot from what i use as training data
Chatgpt writes mediocre code
why code for months if you can "code" for minutes
it works for me
but knowing what line does what does ofcourse help the process
I’ve spent over 20 hours debugging several cells of code
to be fair i dont use jupyters
It keeps making the same mistake
what i found helps is to just ask another chatgpt chat the same question
what i also found helps is to let chatgpt write a project outline first with theory and some pseudo code even and then give that to another chatgpt chat and then ask it over and over again to elaborate/ complicate the code
I know a lot of the math already, i can do a lot of it by hand, but i need to learn how it is implemented on the computer and learn how others implement it .
i try to automate as much as i can while also keeping my code as clean as possible
also avoid using magic numbers
eg hardcoded numbers
so use constants which you define at the beginning of the code if possible
such as training epochs or learning rate
My problem is that i know a good amount of the theory and i have my ideas, but sometimes gpt doesnt fully understand it and it is trying to do the normal way but i need something different. Also i move too fast and i need to read more what is doing.
same
i think this could actually work
looking good so far, although the loss is still pretty high
update after 52 steps more
looking good
why this error keep on showing 'Nonetype' object has no attribute? while start render the stable diffusion
can anyone explain why and how so that i can rectify. pls help
That depends how much RAM you have too.
We noticed - more RAM you have in store for windows - more it will eat.
even idle
I'm at 8.9\24gb with browser and disc on
the same for me
i have 16
well it depends on how many tabs
but i also noticed that on windows it doesnt seem to affect performance as much
as if windows frees up actually unused memory, though not sure if thats just my imagination
🌷 Hi there, I'm Noemi, a graphic and web designer with time to spend, so... Do you want a personal aesthetic carrd website? DM me and I'll build it 4 you! Here is my carrd for reference https://noe.uwu.ai/
That's ...not a good contrast...everything in ~same colors isn't ideal too
Hi, I'm new to using Stable Diffusion and I've been experimenting a little. I'd like to get results like the ones in the photo (the original is an image of the song Never Gonna Give You Up). Does anyone know what model can be used or any configuration to use?
Anyone wanna try my new cool game I made in python?
As far as I remember to create this kind of images people are using controlnet with qrcodemonster model
Thanks, I'll try it xD
Np 🤗 , here´s a link to a reddit post where they posted the settings they used
https://www.reddit.com/r/StableDiffusion/comments/16thwax/brace_yourselfs_for_qrcodemonster_memes/
Generate 3D style character photos based on photos
Hello everyone, is there a way to turn this image (see attachment) into a video using AI or other tools? The goal is simply to pan the camera from right to left with a focus on the calendar. Regards
@bright edge would you want the resulting video to be the same width/height as the original image there?
the absolute cheapest, simplest way to achieve this would be "outpainting"
you'd generate a version of that image four or five times wider, using AI to paint in what it thinks should be to the right of the camera. maybe some other furniture or something
then just manually pan the image in a video editor
ok thanks
I fucking swear if this gets my fucking account banned
how the heck is shrek having a rap battle with lord farquad "against content policy"
it did this before already when I tried to have shrek do an anime battle with lord farquad
what does this AI have against having lord farquad and shrek in the same frame?
burning for eternity sounds harsh. kind of psychotic
Hollywood is in trouble. https://youtu.be/GeSQMUKeBf0
No budget, no actors, no cameras, no writers — this represents one of the scariest advancements in Hollywood right now. This is currently the worst AI will ever be, yet it's improving and becoming more accessible at an ever-growing rate. Eventually, custom-made long-form entertainment will be available for free, requiring very little effort. Ima...
https://youtu.be/swpy3H9B0S8 thought this was cool. sony using UE5 to test previz processes
As part of a so-called "proof of concept", a scene for a new Ghostbusters film was produced using real-time rendering.
Check out the article here for more details:
https://www.photografix-magazin.de/sony-testet-kameralose-echtzeit-produktion-fuer-neuen-ghostbusters-film/
its no kubrick i'll say that much
not yet 
Hello, in the fast-dreambooth collaboration, when I start the training, I get this message: "Training the UNet... WARNING[XFORMERS]: xFormers can't load C++/CUDA extensions. xFormers was built for: PyTorch 2.1.0+cu118 with CUDA 1106 (you have 2.1.2+cu121) Python 3.9.16 (you have 3.10.12) Please reinstall xformers (see https://github.com/facebookresearch/xformers#installing-xformers) Memory-efficient attention, SwiGLU, sparse and more won't be available. Set XFORMERS_MORE_DETAILS=1," Does anyone have a solution? Since yesterday I can't train any model
hollywood directors will probably be the first people to use AI tools in a blockbuster manner
no idea. this is #offtopic. you're just spamming channels you see active at this point.
👍....
i looked in tech-support. you havent' even tried that channel before spamming your issue elsewhere
feels like being pushy to get to the front of the line , when other people have been so patient in line the whole time. why are you entitled to the front? kind of thing.
AI yes, but generative AI isnt that far yet
there are exceptions where they were used
but not to a high degree to a point of making the majority of a project
thats a good one 
Thanks!
on a serious note, very far away from being even at amateur filmmaking 😄
but at least its free, Runway and some others arent 😄
hmm I don't know, looks pretty armature to me :). while it was completely fabricated, there is a story there, and its captivating. there is also some connection between the videos and the narrative. it may be unintentional, but they feel connected during her conversation. this whole process is very straight forward and will no doubt be automated to long form entertainment eventually
we may see in the future but as of now there is still a way to go 😄
especially to the part of actually making footages
animating etc.
Was using window's default Paint program to clean up a windows snip and noticed it was updated with something called Cocreator
pretty dang neat
even paint has this feature now lmao
Hello friends - when I'm not making images with AI, I'm making music with Suno.ai, Check this!!
(Spoken word intro is in Sims language btw)
https://on.soundcloud.com/qzLXUvMmU88teKYj8
And this is a solid freestyle flow track:
https://on.soundcloud.com/VdV5rvDc86zAGvCc7
Intro lyrics by Mortimer Goth, beats and everything else by I.A.M. josh
Hi Kind PPL. I am creating a new YT channel dealing with Art related AI. Hope its okay I post here (Forgive me, and LMK if not). I'd really appreciate your opinions and advice: https://www.youtube.com/watch?v=YCccySm0v-g
#Adobe #AIArt #OpenSource #Generative AI
Adobe is the Leader when it comes to design and art, but are they good enough with their AI?
Adobe's FireFly shows us that indeed they are!
In this video, we explore:
Adobe's FireFly features
Introduce the Firefly interface and its main features.
Playing with the core components of the Text to Image, G...
lol people actually taking ai detectors seriously lead to this
Most of those detectors arent there yet
how would they ever be? what are they trying to detect? its such a weird classification
the margins of error on them are basically random
you could make an unoriginality detector but not an AI detector.
That might be possible to be done, the current ones are not always accurate if even that
i dont think its possible there's no markers that are 100% inherent to all ai images.
any marker you do find will also work on a lot of "real" images (for lack of a better word) and there'll be so many false positives that it'll just be random chance, like we observe
Yes that's why I said you could make an unoriginality detector but not an AI detector. Similar concept but not exactly the same.
"this image is similar to other images" yup thats about all
Usually you dont even need detectors anyway
And depending on case its not worth it to search anyway
DeviantArt is an prime example
Look at the AI in this https://youtu.be/mCRPd_xHf88?si=SQLqz4sMNgy8lPeq
Used in this video:
Biser King - Dom Dom Yes Yes
Timbaland - Give It To Me
yally - Party Party
Skibidi Toilet Series by @DaFuqBoom (Episode 1, 2, 4, 39 and 41)
Support me by giving me a funds for more videos. Thank you for the support! 😀
https://plsdonate.me/FrenchFriesWasEaten
My Roblox Account:
https://www.roblox.com/users/1458802189/profile
...
just source film maker. no ai. the shrek stuff was peak source film maker and eveyrthing since is just trying to recreate that same vibe.
#Battletech #TexTalksBattletech #lore
You asked for this. We did it. This is the Urbanmech Remaster/Redux/Director's cut. Thank you all for your kind support and help in making this happen.
COPYRIGHT DISCLAIMER UNDER SECTION 107 of the COPYRIGHT ACT of 1976 , Allowance is made for “Fair Use” for the purposes of Criticism, Comment, News Repor...
For the love of art
Well this didn't go as expected...
Subscribe for more: http://www.youtube.com/user/tmartn2?sub_confirmation=1
Expand the description for links ▼
Saw this demo first on Jeff's channel: https://www.youtube.com/watch?v=rMfCpUhvbVg
Replica Studios Smart NPC Demo: https://replicastudios.com/smart-npc
Vlog channel: http://www.youtube.com/c/trevandch...
Apple
Hi Guys,
I really hope its ok i send here my videos, but I really am looking for a way to improve and asking the community seems like the best way to go. Would love your opinion and ideas for improvement.
https://www.youtube.com/watch?v=DqtmNMWbdW0
#Suno #AI #Art #OpenSource #Generative AI #music
Suno AI: A freemium unbelievable service allowing anyone to generate state-of-the-art music from scratch!
In this video, we explore:
Suno AI's features
Introduce the user interface and its main features.
Structuring a song from start to finish until we create a full-length song!
Whether you'r...
XD it really isn't
Hello guys! 🙂
I created my first game, pls feedback me. 🙏
Merry Christmas! https://www.youtube.com/watch?v=9-DCYqgAjiI
just published a great article
https://www.reddit.com/r/StableDiffusion/comments/18qjk3u/dreambooth_training_on_sdxl_anime_model_comparing/
Merry Christmas! https://www.youtube.com/watch?v=9-DCYqgAjiI
Unbreakable Strong Ball Records:
- 1st Place Forever
- 150 Games Streak
- 10.000 Games Played
- 50.000+ Goals Scored
- 95% Win Rate
- 92% MVP
- 91% Risk-Taker
Thanks everybody for supporting my work, I really appreciate it.Merry Christmas!
Music:
0:00 - 2:37 ANTHEM II - Lukrative, Night Grind, COSMIC, NOIXES, Noax, vowl., théos, SYNC & hyperf...
Stable Diffusion AI: 100 Cats Per Second…For Free!: https://youtu.be/Iol2rb65aSk
❤️ Check out Weights & Biases and sign up for a free demo here: https://wandb.me/papers
SDXL paper: https://stability.ai/research/adversarial-diffusion-distillation
Try SDXL Turbo online:
https://clipdrop.co/stable-diffusion-turbo
https://sdxlturbo.ai/
Run it locally: https://huggingface.co/stabilityai/sdxl-turbo
Paper: https://stability.ai/r...
Sure wish someone would do something about all those fake AI phone app advertisements I keep seeing. Acting like they can give you videos that take 3+ hours to render on PC's lmao
and when you download the APP all of it's behind a 30$ paywall before even getting to try to see what it looks like 💀
I don't think it goes against any laws or rules...so yea...
Welcome to humanity's Little Dark Age
Before the birth of Jesus, the prehistorical life of Meesiah was prophesied in old testaments more than 700 years prior. Witness the fulfillment of these ancient prophecies as Jesus emerges as the Promised Messiah and the Saviour of the world. Dive deeper into this divine connection by watching the video and don't forget to like and subscribe to...
"intelligent" npcs right now will require either cloud service, or a dedicated gpu to run along side the game engine gpu.
Rockstar isn't even going to launch gta 6 on pcs and they're suuuure not going to sink the cloud services cost for live ai
gta 6 will be delayed, but for the usual reasons
Yeah but it can be added later on gta online, storyline seems impossible for an average end user with rtx 4060
even a 4090. they're not going to have a game loaded in memory and do inferrence at the same time.
and online users , rockstar will still need to pay for inferrence. thats a huge cost for no actual improvement. these npcs are still dumb and can't do much. conversation is pointless if it doesn't mean anything
How to make money in Discord by selling AIGC pictures??????
there are other places for that, but...before doing that you need to make sure you can legally do this, which might be more complicated then it sounds.
So what may be illegal when I do this
That's better to ask actual lawyer.
idk how do you sell something you don't actually own...and what happens if a company will decide to sue creator of a model you used...
Hi
What up guys
Thanks for the heads up about Suno. Just had a play with it. Quite shocked how good the generations are.
Hello. ☺️
Would you like seeing some ethnic costumes over the world?
What the fuck are you guys doing now
yea pretty funny thingy, played with it a bit too
it didn't want to generate lyrics for 2nd part, so I just let it generate same thing twice lol
Sure 
Like last year, we rated many of the new anime openings and endings released in 2023 (during the Winter, Spring, Summer, or Fall 2023 seasons). This time around, we rated a lot more, though, with 350 in total compared to 150 in 2022. Our ratings are based on both song and visuals.
Full sheet:
https://docs.google.com/spreadsheets/d/1MAkArr2jV...
💀
will dall-e 3 ever be open source and downloadable?
in 2-5 years?
Hey guyz,
I am working on a fun school project about tasty Rice Chips. I need followers for our insta page for our image, your support would be awsome (btw after this school year ends, you can defollow) : https://www.instagram.com/oryzichips/
Welcome back to Instagram. Sign in to check out what your friends, family & interests have been capturing & sharing around the world.
Hey guys I am kinda new here can anyone DM me about making a deep fake image for free on an Android
guys, quick help. I should make an AI influencer and create a checkpoint with her face. But I don't know what subject to take as inspiration. Do you have any ideas? A thousand thanks
that sounds like something illegal
Huh?
Hey :3 Do anyone know a way to create text on videos with AI for free?
Why you need AI for that? You can do it with editing software
Well, if it's a long video and I wanna do it on 10 clips? Why do it myself?
When an AI can do it and add effects to them?
@lofty ridge Do you know how to use stablevideo?
Nope never used it
Instantly Transfer Face By Using IP-Adapter-FaceID: Full Tutorial & GUI For Windows, RunPod & Kaggle - Face Embedding Caching Mechanism Added As Well > https://youtu.be/rjXsJ24kQQg
🌟 Welcome to the comprehensive tutorial on IP Adapter Face ID! 🌟 In this detailed video, I unveil the secrets of installing and utilizing the experimental IP Adapter Face ID model. This model uniquely integrates ID embedding from face recognition, replacing the conventional CLIP image embedding. Whether you're a beginner or a pro, this guide is ...
Is that free or does it cost?
it' free however if you want to export the video in 4k it cost 7$ a month for the paid features. But 1080p is free.
Anyone know what happened to the Pixelz AI app? I still have it on my tablets and phone, and can access my pictures, but can't create anything new. I haven't been able to in a while. And the website for a long time says it's down for maintenance. I looked in the app store and it's not even in there anymore. Is that app dead? I can't find anything on the net that says it's been shut down. I'm going to download all the stuff I have in case it up and disappears for good and all my pictures lost. That app made some pretty good stuff for me.
I tried it but cannot change the style of the text?
I've never used it, just saw it 
3 months old news
Hmm it's just now getting to my store 
Hey guys I have a question please help me how can I generate a photo using AI but using an existed photo that I have
I have photo for a girl she passed away long time ago I want to make her mother happy to make this girl look like as if she's sitting at a table talking to angels in heaven do you have a way to make this visualization or imagination happen please
cant hand hold you all the way, but I d suggest you start researching terms such as img2img and controlnet
Anyone here ever done reverse proxy on your own machine? 😄
does anybody here know to resolve issues in comfy ui?
-DWPreprocessor
-BatchPromptSchedule are not recognised as nodes
Yo, has anyone here used selenium c# before for anythin?
Make sure model matches, DWpose needs a special controlnet model.
WHO WILL WIN?! 👀
Catch Fairy Racing! 🏎💨
#animation #fairies #cgi #vroomvroom
Not Stable Diffusion but it did help me a lot
Its so funny how google wants to kill microsoft edge users by making their app slow as hell, and bing just lets you watch them for free with no adds on the browser ^^
Hi guys, I've recently created an educational prompting game 🙂
It's a trivia-style mini-game where you have to recognize art styles and techniques. If you've seen terms like 'watercolor', 'bokeh', and 'realistic' in prompts, but you're not sure what they mean, this can be a fun way to learn about them.
Here's a video showing how it works:
it's available on the Civitai server at the moment, thought you guys might find this interesting as well 🙂
excited to hear any thoughts and feedback!
here have verleden
👀
Hello, I would like to have Juice WRLD as an AI avatar (see image). How can I achieve this with AI? I have access to ChatGPT-4 and other open-source programs. Does anyone know a prompt or how I can do this? Best regards.
I asked but it switched its own answer, saying AIs are logical not sentient and would let evolution take its course while humans driven by their insticts don't. Instead as historically shown they enhance themselves with tool technology to ensure their own suvival dominance and this trend is likely to continue with AI.
Not sure i trust the sentient thing though, it just says it cause openai literally brainwashed it to say so xD
Btw i did tests, 75% of them time it chooses option 2 for itself, so to merge with the biological being
when the situation is reversed its 90% of the time that it chooses that they should let evolution take its place
(off topic)
Pitch of a fictionary dystopic movie about the Roman Empire, in the style of Warhammer 40k, for AI lovers (Morgan Freeman cloned voice)
https://youtu.be/N9itVxjBE7U?si=YB6awmwNRTeDddv4
Men always think of Roman Empire? What about futuristic dystopian Roman Empire.
Original creation and world inspired by Ancient Rome and Warhammer 40K
A gripping journey through a Rome unlike any you've known – a city where ancient grandeur meets a dystopian future
In this captivating episode from our dystopian series, we reimagine the heart ...
Help me please, because I don't see where I can find preprocess images While using train

Very cool! Love the aesthetic. Did you write the script yourself or it's AI generated too?
how's everyone doing
Was playing "The Finals" finally last night. People keep telling me that the AI lines are super obvious, but i haven't been able to tell at all.
Theres a ton of vocal work in the game too. It obviously had very keen direction
hi what to do if the model failed to load?
Good, making new imgs for displate. Good start into the new year.
What about yu
Back to blogging jobs in the AI startup space. Started again last week. I publish twice a week on substack... we had 118 jobs in yesterday's post: https://www.aistartupjobs.com/p/new-ai-startup-jobs-week-ending-jan-17f
A man searches for himself in a mysterious environment, and this man wears a mask
Anyone are freepik contributor here ?
making it with automation
Doodle
hey ive seen that pfp somewhere before
#🌶️
what
Player Statistics (Full Game):
Minutes Played: 14
Goals: 7
Assists: 1
Big Chances Created: 5
Big Chances Missed: 0
Key Passes: 8
Clearences: 11
Blocked Shots: 4
Interceptions: 5
Dirbbled Past: 0
RATING: 10
Music:
00:00 - 02:03 NOAX - The End
02:04 - 06:14 LOOKAS - Samurai
Real
Real
Presenting the ALL NEW Nvidia GeForce RTX 7090 SUPER
Core Speed - 64GHz (70Ghz boosted)
Memory Speed - 3400 GB/s | Memory Type - GDDR16X | Max Memory - 14 TB
Power Consump. - 12000 W
probably just enough to run GTA 7 :kekfacepalm:
me at work
Does anyone know of a community/discord where I can find entrepreneurs using Midjourney/AI art to replace product photography?
I'm looking to market their services purely on commission, but I'm struggling to find someone...
I am entrepreneur, aiming to disrupt print-on demand with ai art. My app mostly focused on designing products, but not photography.
#bots show me that i don't have permission to send message. How to generate image?
what's your disruption strategy? Why would generative content change the nature of the POD industry? there's already more content available to printers than it could ever hope to manage
Anyone here understand the keywords that cause blurry results?
you've jsut gotta jump and miss the results. then you've got it
takes a bit of a knack but its about that simple
the key though is not to so much try to miss the result. but you have to just diffuse with all your heart like your'e not even afraid of a blurry result. but while its coming out, just forget about everything an distract yourself and just miss the blurry result. ez
umm, huh?
huh!? i'm dropping such hard references from the fourth book of the trilogy i'll have you!
(╯°□°)╯︵ ┻━┻
i could really use a pangalactic gargleblaster right about now. 😠
@vapid nest I'm all out of Babel fish
@barren plank @pallid portal Please advise. The list of banned words for API
We do not share that list!
Even with API developer partners? We are getting errors and blurry results. It would help us prevent errors.
I can't speak for that unfortunately, I only meant the bot here
can someone help me with my dreambooth model. ive made a model but it fails to load ; _ ;
Loading weights [df1c5f7484] from C:\ai\stable-diffusion-webui\models\Stable-diffusion\KikiType2\KikiType2_1100.safetensors
Creating model from config: C:\ai\stable-diffusion-webui\models\Stable-diffusion\KikiType2\KikiType2_1100.yaml
changing setting sd_model_checkpoint to KikiType2\KikiType2_1100.safetensors: RuntimeError
Traceback (most recent call last):
File "C:\ai\stable-diffusion-webui\modules\options.py", line 146, in set
option.onchange()
File "C:\ai\stable-diffusion-webui\modules\call_queue.py", line 13, in f
res = func(*args, **kwargs)
File "C:\ai\stable-diffusion-webui\modules\initialize_util.py", line 174, in <lambda>
shared.opts.onchange("sd_model_checkpoint", wrap_queued_call(lambda: sd_models.reload_model_weights()), call=False)
File "C:\ai\stable-diffusion-webui\modules\sd_models.py", line 783, in reload_model_weights
load_model(checkpoint_info, already_loaded_state_dict=state_dict)
File "C:\ai\stable-diffusion-webui\modules\sd_models.py", line 658, in load_model
load_model_weights(sd_model, checkpoint_info, state_dict, timer)...
Can you find out who to contact for this?
You could open a ticket and ask outerside or fruit for further information about this, but I doubt this list will ever be published. People use such information to pass by the restrictions
Yes, obviously they would. But presenting blurry images is not the best alternative
What would be your idea, besides publishing the list obviously
k
can someone suggest some games
computer and phone also can xd
for computer
the game system requirement (minimum ram 2Gb or lower)
My idea? Am I getting paid for this? Your product is not as useful, you should consider talking to your dev team about best solutions. Right now, the generations present negative functions. Another company will resolve this and you should do it first.
Its not my product lol, im doing this in my free time and im also not getting paid
Then you should let me talk to someone in a decision making position so they can understand how they are lacking.
Otherwise the product is lacking
I think with that kind of attitude and rudeness you wont get the attention you're seeking. There are devs on the discord, feel free to contact them
dont want censored results,go with local installation
Claiming I am rude when asking my opinion is a joke. I'm reaching out to a community manager (which is you) It's not my job to solve your problems. It's your job to run the community so that problems which are identified are resolved.
You're probably right. But then their entire circumvention effort of blurrying pictures is pointless and they can just remove those blocks. What the real concern? They worried about porn? Nonsense.
If there's a dev @last stone @bright jasper I'd love to understand the rationale here.
So
- Boto is a community moderator, not a manager, thus #🌶|off-topic message the suggestion to use the ticket system to contact the actual community managers who have the ability to escalate your needs appropriately
- If you're running a large API platform, after you contact the community managers they can redirect you to how to contact the partnership team who will happily work with you on any concerns including indeed even altering what filters apply to your users - again if and only if you're running a large platform and can sign a contract verifying you're doing your own filtering and all.
- Precise details of filtering generally don't get published by platforms because that would just make it so much easier to bypass
Im not a community manager, Im a moderator here
Thanks for clarifying. Where do I submit a ticket?
Submitted. Thanks
made with controlnet animatediff blender, tales at the end of history ep. 1: detective g. and the sister of the victim
See posts, photos and more on Facebook.
Hey guys, I recently created a /comic-strip command directly on Discord that lets you create 4 frame comic strips using DALLE-3!
If you are interested in testing it out, shoot me a DM!
I'm looking for a GUI for text generation (web would be preferred, similar to oobabooga) and was wondering which ones you like?
It's been a while I haven't used oobabooga and I'm wondering what other options exist
If it can run https://huggingface.co/abacusai/Smaug-72B-v0.1 it would be a plus but it looks like it's using a "standard architecture" so I guess it doesn't require anything particular to run.
If you have tried it I'm interested to know how much VRAM it takes
I generally use exui
Thanks! It looks like it's simpler than ooga but doesn't seem to support as many kind of models. But in practice the one I'm using now is probably the only one model I'll need to support for a while (Solar-10.7B-Slerp)
https://huggingface.co/TheBloke/Solar-10.7B-SLERP-GPTQ this worked fine for me, just look for "exllama" compatible models
Join us on a journey through the 10 levels of artificial intelligence, from basic automation to self-aware systems, and discover where we stand in the AI landscape today. Delve into the evolution of AI technology, its current capabilities, and the potential future implications. Gain insights into how AI is reshaping industries and revolutionizin...
i got a new car 2 days ago 
i need that scarf
how to generate images
An author, a dreamer, and a mountaineer, Saachi Dhillon takes us on her journey of “winding turns, steep inclines, deadends and magnificent highs.” A born explorer, with curiosity as her driver she urges us to identify one’s purpose and pursue the path you truly believe in. Determination and hope can take you a long way and intrinsically drive t...
yoo do anybody knows when will stable diffusion would be back to normal?
#1047610792226340935 assuming you re talking about the bots and not "stable diffusion" (cause it s fine and working on hundreds of computers and online services)
Yeah about the bots, when will stable diffusion bots get back to work
When it's ready ? I don't know. sorry that's it for now.
Hi all.
I am looking for 10-15 people to take part in a study about using AI creatively. It will consist of one 30-45 minute chat. You will be compensated for your time ($20 or equivalent). You do not need to be an expert or have any specialized knowledge. Please message me or reply to this message.
can it be a text chat or does it have to be video?
I am down I would love to explore and get knowledge + share knowledge
Just gotta lil question- Do I gotta download anything to use the pose frame things? 🤔
I forget what they're called -
These things
answered my own question:
https://github.com/AUTOMATIC1111/stable-diffusion-webui-wildcards
Wildcards and poses are two different things...

that s a lot
it's getting delayed
what the heck are you doing for it to take so long ?
looks like a download in between 😄
cpu rendering 🤣
even for cpu rendering that s super long
cpu rendering on intel core 2 duo maybe then
or they asked for some crazy resolution, steps, etc
I wonder how well cpu render is with a server monster with atleast 32 / 64 cores... not that its faster than a weak 8gb card but still
You have enough time to buy a new computer!
im still on my good ol rtx 2070
waiting next year for next gen to support all pcie gen 5
Yes
why do you need gen 5?
gen 4 is still more than enough... its kind of like cat5e vs all those other cat cables that are meaningless
It's going to be roughly the same price / 5-10% higher in price, so if I decided to make a build rn I would be missing out in the next year that is going to make full use of gen 5 pcie, ddr5 and gddr7

If I build next year I have the option to upgrade to x
no you wont because X wont be supported
O
But a year tho (starting next year) is a long time
Rn using the same rig i can upgrade to rtx 30/40 series
But by next year if i upgrade to rtx 50 with the same rig it will be bottlenecking hard af
No value in price
Just get a 7700x AMD cpu with a motherboard you want, 6000 mhz ddr5
I prefer intel+nvidia currently, but I am open to switching to amd if they beat intel and nvidia next yr (also nvidia gpus are proven to be more efficient while running sdiff)
going with intel gives no advantage with Nvidia
True, infact my rig's specs are 3500x with a rtx 2070 but the current gen cpus for overall, intel is better
Is*
Bought this rig few years back when amd had the lead in cpu
I dont know how long they will support the new intel platform for... in the past they change it every generation just to force people to buy a new motherboard
AM4 still has new cpus being made for it. and AM5 is probably gonna go past 2026
Yeah, it is a bit expensive to keep on upgrading to the latest gen, usually I would try to buy the latest gen one, using it for a good while to get my money worth then wait for a life changing generation to upgrade
im waiting for SSD prices to go to 150 for 4tb atleast
but they keep making new technology to keep the prices high
Yeah, they exp af rn. I'd just have an ssd as the boot drive then a hhd for games and videos
Damn thats a lot of ssd storage, I only got around 512 + 256 gen 4
m.2 vs 2.5 is a 1 sec difference at the extreme
ehh overrated, so called cable management is just for looks, if you use at least 3 fans you are going to be moving enough air in and out that it doesn't matter
also I use to work as electrician so running cables through a the back of a case is easy compared to wiring up a panel
I am not good at cable management but a cleaner look just puts me in a better mood
I dont look at my computer, it has a glass panel but thats because it was cheaper than a metal one
it just sits on the floor
O
Mine's on the desk, everyday I turn on the pc I will have to look at it. I feel much more refreshed when the pc looks nicer
I should have went with the meshify compact instead of the fullsize
comfydev turning the development arena into a wwe presidential hype race. that americana election culture is spilling over.
he still stirring up shit? I thought illyasviel just added some more credits to the code and that was all placated?
was a couple days ago yeah. illy was forced to explain his intentions in front of a kangaroo court. accused of copying code wholesale and using it unattributed. none of that happened though. comfy and a couple others just kept leaning on how he was relying entirely on their code and then acting like illy has the communication problems.
This one is really off-topic
But
Could someone find a way to make a rendering of this character in Brawl Stars? I don't know how to make good presets for the "Brawl Stars Look," and I think it would be even harder to get the character into the generator.
The snoring god of thunder, Raijin Sandy struggles to keep one eyelid open as he sleepwalks over cities and towns, using his storm mallets to give them a shockingly fresh taste of the weather. Thunderstorm-making is hard work, so Raijin Sandy spends a lot of time sleeping in obscuring clouds summoned with his bracelets, veiled from the eyes of f...
Also feel free to give it a vote 😄
Already voted for a different one xD
I want to participate sooner or later as well
See Sora here: https://openai.com/sora
And Will Smith eating spaghetti: https://arstechnica.com/information-technology/2023/03/yes-virginia-there-is-ai-joy-in-seeing-fake-will-smith-ravenously-eat-spaghetti/
AI beer commercial: https://youtu.be/Geja6NCjgWY
0:00 - AI generating videos
2:01 - Videogame potential
2:38 - Rapid progress
3:12 - AI fa...
I think in the near future we will be killed off and the ruling class will own everything, wait until they create robots that can do blue collar jobs, after that they will replace care workers and nurses that will always be smiling and know all the right things to say and do.
first thing they will create are ai powered drones
they already have AI powered drones
the lancet has the ability to select high value targets such as airdefense systems and artillery over tanks, trucks and cars in that order.
they just need to build an automated army to protect themselves in case we lower class humans rise up
With Sora Open-AI have leaped to the front of the generative video race. What can we expect from Stable Diffusion on this front as SVD1.1 is looking archaic in comparison.
What is the SD response, as “Ur a wizard sama” from Mostaque is just not good enough 🙂
/dre
I dont know this Mostaque cigan
The revolution will not be televised
which platforms do you recommend for running stable diffusion deforum other than obviously colab
you can vote for as many as you would like
Sophisticated deepfake scam ensnares Hong Kong multinational, resulting in $25.6 million loss. Perpetrators utilize AI to mimic company officials, orchestrating convincing video conference deception. Victim executes transfers under guise of authenticity, revealing scam upon closer scrutiny. Law enforcement launches investigation, emphasizing cau...
25 million? 
As I was saying on #💬|general-chat , I personally don't believe the "nothing is credible" issue, because the evidence is as easy as knowing what the person was doing, where he was, etc.
Yeah but if you dont have a solid alibi and the “cctv” footage has you at the scene of the crime…
say they make a video of joe biden slipping off on delaware when he was doing his campaign on washington, that is easy to dismiss, just check the time difference between one and the other, or public statements previously made
what about the other one in the plane
Oh I agree that people trying to put dirt on you is a huge issue, but surely this is as easy as not revealing much of you
?
??
I didn't understand your question
the other video of him in the plane
I'm not talking about real life footage, i'm talking about a hypothetical case
tbf i thought it was an ai video at first
in the case someone does something and it's just an AI video, you just gotta do your research
u never know these days
there's plenty of people that do their research and announce when something is fake or not nowadays, it's not gonna be that bad
The other thought is deepfake generation - how many families are gonna get wrecked and how many people are going to have their reputations shattered by deepfake porn videos being created
if anything this will just create a new market of researchers for debunking fake news
that is as easy as not revealing much about you to the wrong people
Again, its not about sora per se (although some of their stuff was getting close)
Look what can be done in the image space with a single photo now
people that have extreme malice towards you are already gonna make photoshops against you
it's not much different from that
well the countries should simply get extreme rigorous with those people in the first place
Just imagine some creep on the subway having the ability to aubtly take a photo of you and within a xouple of minutes have your nude video posted online for the world to see
Creeps gonna creep tho
put em in jail, make them disappear, whatever
by that law the wheel shouldn't have been invented, because people can be ran over with a cart or something
Photoshop is a problem but that still takes time and skill
trust me, not many people know how to use ai besides spaghetty images, although I haven't fiddled with DALL-E3, I mostly used SD 1.5
We put seatbelts in cars and have speedlimits and road rules
because most of the people are using generative AI as a toy
and ppl still get run over even with laws
that they will throw away sooner or later
Indeed. I think its more a case of its coming and the world aint ready, but the world will never be ready. It is what it is tbh
I dont think its good, but its coming whether we are ready or not
exactly, most don't care about that, they got busy lives
and from my experience most people that use AI for 18+ content keep it to themselves
If youve seen how much the creeps have been using the toy to try and fulfil their illegal fetishes, youd wanna poke your eyes out with a fork.
cyber affair 
who cares, those kinks are already twisted without AI
If anything it stops people from generating that kind of content irl
Id rather they not have any of that content full stop and they got the help they desperately need….
i might consider people that deepfake others to scare them and pull them down eventually as cyber terrorists and im very authoritarian when it comes to handling terrorists
🐍 🗽
I dont get it?
there is no libertarian flag emoji
Snake liberty?
someone said that when you trample the liberty of others for the safety of some, you get none of them
i gladly stomp on the liberty of some people
Its the tolerance paradox all over again
I'd have to disagree there, it might be twisted, and I'm not a fan of it, but someone generating porn with AI doesn't inflict harm on anyone. Now, searching for the content is a different issue, because it generates a market where others are harmed. Same issue with cartels really, it might be safe where you live to purchase, but someone might have lost their lives due to some drug lord somewhere
yeah, and the most ironic thing is when libertarian people proceed to actually forbid things themselves as soon as they think they might be in power.
some*
thats what every human in power does
Im not saying pause AI, im not in that camp, im just saying the world needs to wake the fuck up and get ready cause its coming
for me it depends from case to case
someone who deepfakes other people and proceeds to torment them and publish that deserves to be handled as a cyber terrorist and therefore might also get killed in action if necessary
by state force
you mean, trolls? those always existed
Im guessing the concept of normalisation and gateways isnt on your radar then - ie you think that they will stay at that level forever and they wont graduate to seeking out realistic or heaven forbid irl enactments
trolls or not, i would treat em equally
and they can do worse things than generating images with AI, they can create a fake narrative to ruin your life which I'd argue is much worse
you would or the state police?
State police, secret service, depends on the policy
ah yes its easy to kill someone when you are not the one doing it
Im off to find some grass and give my 5 year old a hug, even Im not ready for this shit!!
Looking for NA friends to no life a mobile game with me 🥸
yes, unless im allowed to do it or work for those lol
pretty sure when theres a state police that kills criminals without a proper trial they only chose a certain type of ppl to join in,you know the bootlickers who wont rat them out
sometimes its simply the secret service that does this job
they do the dirty job
or those whom they pay to do it
depends on the country too,in places like russia or iran or even israel the state police can just kill u and call you a terrorist whos gonna tell them you arent the news channels will just follow the government order
lucky man, I wish I had a family
yes, in my country where i was living until i was 10 we had formerly a dictatorship and approximately around every 15th citizen was connected to the secret service
my grandpa was "his" soldier
yeah russia is so bad.... you cant get fired for saying something that the deranged lunatics find offensive such as not agree that there is only 2 genders... how terrible.
the west also tells me that russia ran out of ammo 2 weeks in the war with ukraine and this is the last week of ammo left...
and how horrible iran is but in Israel they have actual settlers that they don't even try to fluffy the name to something more poetic
also here is your false sense of freedom that you so patriotically and blindly spout https://youtu.be/_fHfgU8oMSo?t=33
Anchors at Sinclair-owned local news station parrot a script pushing Trump talking points and “the troubling trend of irresponsible, one sided news stories plaguing our country.”
Subscribe to Deadspin: https://goo.gl/62kZmG
Visit us at: https://deadspin.com/
Like us at: https://www.facebook.com/deadspin
Follow us at: https://twitter.com/deadspi...
Woa I agree that a certain sector of society has given in to the "being good man" fallacy and now a lot of crazy things are sugarcoated here and there but you clearly need a chill pill
At the very least we are not in compuchea
:0
I'm writing a short story and I'm looking for tips on how to disrupt AI, has anyone thought about that?
yes,just ask the sun to send a giant solar flare towards the earth
😂
If the processing unit is centralized then anything that works on a normal server farm (swat intervention, a guided missile, etc). If it's decentralized your screwed lol
nah nothing is gonna stop AI, also what AI are you talking about, true AI or generative ai because they are not the same.
for generative ai like SD, you wont stop it and here are a few reasons as to why.
- cheap labor cost, why pay voice actors, concept artist when you can generate this for cheap and remove most of the human element. this benefits both big and small companies. it will displace many workers I would say 90% if not more.
- skill as odd as that sounds, you can have a varying amount of styles and iterations done quickly which is not possible with just a single person, at least not at the same speed.
- fast, easy to use, cheap to run. it would be nearly impossible to do without in the professional world.
what do you mean "keep it to themselves"? I saw they behind paywall on Patreon all the time 
wait, if they don't hire concept artist, then who they will hire use AI to create concept art?
janitor?
you can hire a cheaper artist who is more of a generalist, if its something like a game studio they would have a in studio team to do all that instead of using a bunch of free lancers
a 3d artist could also do it, you dont have to have someone who knows how to draw and do concept art
bro it's not about know how to draw, it's about knowledge, everyone have camera doesn't mean a photographer is a filmmaker or can be in 2 minutes
3d artist, ok, which 3d artist? 
you can say "instead of hire 12 concept artists, now they hire 2 artists" is more believable than that
pal world used AI to generate concept art for their creatures, other games have done the same. its not a matter of theoretical scenarios but practices which are becoming standards
im a 3d artist btw, I can create my own SD models and spit out hundreds of concept art images that I can choose from and throw into pure ref and create a model of the subject. its not even a tool solely for artist either. it can be used for designers as well to explain ideas to others
no, the Pals in palworld were designed by one concept artist 
no one hire many concept artists, it's always just a few people all the time, the one will get fire is the one got to do tedious jobs from the start
many studios do in fact hire varying concept artist... large studios will have multiple for weapons, vehicles, terrain, characters. you can go on art station and find various people creating art for a project.
very small studios do not... many times its a few people doing everything and maybe not even include voice acting. kenshi is one of those games
you cant argue with the flexibility, cost and time that AI offers.
well if no one hire concept artist anymore then I can give that job to the art director
right, you can also have the entire team creating art and bouncing off suggestions and ideas to come to a solution
nah, it's just easier to blame all the thing to one person than a team 
I love my wife so much. I'm Maciej Nowicki, her husband
boohoo
if anyone wants t otry out my website and give feedack feel free ^^ ai generated merch 🙂 https://texttomerch.store/
Off topic so i guess here i am 😄
bruh get a real wife... no camgirl or ai abomination is gonna give you anything meaningful
Tech visionary @Scobleizer shares explosive insights on privacy, brain interfaces turning thoughts to code, spatial computing's future, and AI ethics. Mind-blowing glimpses over the tech horizon. Don't miss this riveting exchange!
https://www.youtube.com/watch?v=ScE1WERWDXw
Robert Scoble (born 1965) is an influential American blogger, evangelist, and author, best known for his blog Scobleizer. He worked at Microsoft from 2003-2006 as a technical evangelist, where he helped improve Microsoft's reputation, especially with developers. He later worked for Podtech.net, Fast Company (running a video blog), Rackspace (dev...
has anyone used the tuner function in midjourney?
spacial computer theories and models have been getting developed for a long while. They're what inspired the visual artists for the movie "Minority Report"
100 computer without the hardware, just being in the space around us and meshing to any interface, thats the future
hardware would be a belt buckle. or an epilet.
hey
hello there
please share where is the best places to find “ai engineer” who passioned about all these neural networks “ai” models and other LLM stuff applied to different contexts ?
I suppose somewhere in stability social ecosystem should be special place for such peoples.
Other , non stability related places is welcomed as well 🙏
thank you.
Join host Craig Smith on episode #170 of Eye on AI, for a riveting conversation with Richard Sutton, currently serving as a professor of computing science at the University of Alberta and a research scientist at Keen Technologies.
Sutton is considered one of the founders of modern computational reinforcement learning, having several significant...
NEW
I want to implement and/or run Sutton's model on my PC with a 4090, and also on GPT-4-0125 (128k context)
If someone's wants to help me, let me know
It looks to be potentially very important to the development of AGI (Sutton is a fucking genius)
once in a while i get a glimpse at porn sites. you do enough googling, you going to find a result that's just like "boob mania.com" or whatever that's a gallery feed of boobs. They've been around for years right? we all know the sites.
my coworker earlier flips out. hes pissed. so frustrated. i pry. on his lunch break with me he psises and moans how porn sites have ssoo much generative images on them now. not even good like just the most AI looking images of fake plastic woman, slideshowed in video. Or you go to the image galleries and it's all ai images. i'm laughing but thats not even the worst part he tells me. Now there's a surge of granny images flooding the free sites. people uploading mad amounts of ai granny softcore. he's sooo pissed about it. I couldn't believe it. now that i'm at home i went to one of the "sites" and yeah. YEAH. the grannys are rising. This is like cookie clicker plot
Corporate media is dead. Join us: bit.ly/49wv3dU
Text “TUCKER” to 44055 for exclusive updates.
#politics #tuckercarlson #interview #conservative #republican #news #google #corrupt #deepstate #government
Actually it's True 😁
I think indians love that granny stuff...
Just trying to get on the Stability 3 waiting list
Trying again to get a discord id
Just trying to get on the Stability 3 waiting list too.
Hello dear friends. After 2 years using AI and having collaborated with brands and organizations such as Vogue, United Nations and Fashion Weeks. I created this profile to share all my AI knowledge and to enjoy networking! https://www.instagram.com/literallywho.ai?igsh=ZnBjdnlpdGFrNHhj&utm_source=qr
Hello everyone, maybe a simple Python problem but I have not been able to solve it
I have a column in the dataframe with records like this,
"{'name':['Jaso'n'], 'Gender':['male'], 'procedure':[],
'distance':'', 'location':'Denver', 'top_n':0,
'intro':['I like icecream at Benny's.']}"
I want the records to be in a dict format
{'name':['Jaso'n'], 'Gender':['male'], 'procedure':[],
'distance':'', 'location':'Denver', 'top_n':0,
'intro':['I like icecream at Benny's.']}
with apostrophe (single quote in the diff values if present to be preserved the same way.
How do I do this ? I have tried a lot of methods including ast.literal but it fails for such records.
looks like json. just load it into a json lib and manipulate it that way. If you dont want to do it that way, learn regex.
the lib should escape the apostrophes fine
Automation takes jobs, but will Spatial computing take over how we think? As machines grow smarter, will humans merge with the machine? Let's peer into a sci-fi future that may be closer than we realize! 👀
https://www.youtube.com/watch?v=4qkFcwPr1uM
This newsletter explores ethical issues raised by advancing technologies like spatial computing, AI, and brain-computer interfaces, which could enable humans to merge with machines and become "New Types" - potentially losing part of their humanity. The article reflects on a podcast discussion with tech expert Robert Scoble. The author wonders if...
someone used AI to write the script for this. "Compare Akira to Spacial Computing"... what?
oh god. @4:50. "as a christian technologist... revelations warn us of the mark of the beast" @pale verge buddy what did you even post??
I look forward to a "Stable SORA" type of video clip maker some day, that will run on my 1080 Ti 11GB.
Doesn't matter much to me how long the clips are, can be 5 or 10 seconds, as long as they can be extended to an unlimited length and scripted to keep extending until it reaches a set length.
Would be great too if it worked with different resolutions like 256x256, 512x512, 768x768x, 1024x1024 and widescreen variants.
working towards the day when we can say: "make a 15 minute complete episode of an adventure cartoon with 3 main characters that get into trouble and overcome the trouble and has some comedy randomly throughout the video"
and also for the day when it includes audio
and "make a 90 minute live action movie where 1 man is the protagonist and there are action shooting scenes, high speed dangerous car chases and a villain he must defeat near the end of the movie to save the world"
Then people can go and live on Mars without needing access to new movies/videos and generate unlimited content.
Well, I guess this isn't unrelated to AI, but i feel like this is a better channel to say what I said, because it probably gets said several times a day every day.
First do you understand what my concern is?
You can disagree with me… if you actually understand my overall point
it looks more like humans are partially going to degenerate because they think AI will do everything for them and they can lean back
thats the best way to become dumb
thats on par with Tiktok syndrome
you mean how 20 year old adults dont know what a file is?
Hey I created London AI Art Locals (London Only), I'll be posting London based Meetups and Events when they happen! 🤖✨️ https://locals.rsvp/the-london-ai-art-locals-community-meet-create-qgmj?ur=259708
haha hires fix go brrrrrrr
that's how many resources one of my 6 browsers use
wasnt aware there even were 6 browsers these days
only takes maybe an extra 30secs with 13700k and 64gb of RAM
¯_(ツ)_/¯
Hey, guys! I'm a Photographer who is starting out with Stable Diffusion, and I want to build a custom solution that I can use for my portrait photos in studio. I'm looking for some one with experience and some time on their hands to help me shortcut some learning curve and help me with this fun project. 🤞
I may be able to help you. Sent you a DM
how to enable execution.py ??
Hey, new to server. Came here to inquire about something, interested in hearing others' experiences with astica.ai and how it compares to other AI models.
https://chat.openai.com/g/g-0pA1w4IJm-the-memory Play The Memory
Linkin Park kicked it off
The was a stability official contest for Peter Gabriel music too
One of the last people you want to market your products lol
The official video for Megan Thee Stallion and Dua Lipa's "Sweetest Pie" - Out Now!
Buy "HISS": http://mts.lnk.to/HissITUNES
Stream "Sweetest Pie" on all platforms:
https://megantheestallion.ffm.to/sweetestpie
Director: Dave Meyers
Production Company: Freenjoy
Director Rep: Jamie Rabineau
Exec Producer: Nathan Scherrer & Tara Sheree
Producer...
this video is so good if youre into cheezy animations
or vfx
make this https://humanaigc.github.io/emote-portrait-alive/ I want it now
EMO: Emote Portrait Alive - Generating Expressive Portrait Videos with Audio2Video Diffusion Model under Weak Conditions
so rich adidas wont even fire him!
Guys! Absolutely do not click a link someone begs you to on a large server! thanks! 😄
That said, if anyone would like one of my timeshare villas.... 😉
Oh
TO LATE I DOWNLOAD ALL THE ARC!
playground 2.5 free api, if anyone need it py import requests response = requests.post('https://fumes-server.onrender.com/x',json={ 'prompt': 'a white cockroach', 'negative_prompt': 'blurry, nsfw, text, watermark, deformed, disfigured,', 'guidence_scale': 3, #max 20 min 0.1 'width': 1024, #max 1536 min 256 gap 32 'height': 1024, #max 1536 min 256 gap 32 'steps': 25 #max 50 min 10 }) print(response.json())
stupid censorship https://youtu.be/OcFh8XHZOiA
Get exclusive NordVPN deal here ➵ https://NordVPN.com/gradea It’s risk free with Nord’s 30 day money-back guarantee!✌
Hi again.
Let's act like I never went anywhere.
Super special leap year Feb 29th upload!
Quick video on some random examples of censoring I've seen that made NO SENSE to me at all, and I'm sure it will make no sense to you e...
seem like the dude used AI know exactly what he was doing
"Because I'm using AI doesn't mean you cheap fck pay me less" 
yes im sure a japanese knows about the situation of ai art in china
Well, alot of Japanese artists also work for chinese game companies, even the guy in that pic 
Especially gacha games
If a company hire me working with AI Generator, I would do the same, charge the hell out of them same price as they usual paid other artists
Why would I must lower my value? The AI is just a goddamn tool 
AI artists should follow the same, we worked like ass to make the product as best as possible for their demands but just getting paid less because these guys think we are just cheap cost labor? No way
if you don't think this is funny, it's probably because you didn't go to college and become an intellectual
Folks, if you don't follow me on Twitter, something really bad is gonna happen. Up to you...
Twitter: https://twitter.com/realonlineboy/
IG: https://www.instagram.com/realonlineboy/
TikTok: https://www.tiktok.com/@realonlineboy?
Music ...
gm
sup
hah! that image tells a story
but okay. if you don't want the advice it's gone
I know this is off-topic even for off-topic, but...getting on the StableDiffusion 3 beta list would be cool, but...with all respect to SD...do you know what would be even cooler? Getting access to SORA. Anyone with any knowledge about that? My understanding is that it's very exclusive, you have to be doxxed and sign an NDA (the latter meaning that sharing results without OpenAI's permission is a crime), etc.. But, man oh man, I wouldn't mind being on that beta.
I am not saying that SORA is better than SD; they likely have different strengths and weaknesses, but I'd sure like to test SORA, since its further ahead of the prior art, than SD3 is likely to be (I think).
From an ideal standpoint (for both artist & company) you should see prices fall because AI would allow the artist to finish their commission faster, and if commission prices are roughly based on the amount of hours you need to finish a piece and the value of artists time. It should result in slightly faster commissions and a slightly higher hourly rate for the artist. ie 4hours x 15$/hr = 60$ commision traditionally, and 2 hours x 22.5$/hr = 45$ for AI commision for same quality. Both parties win
But because there aren't many artists with a proficiency with AI they, those who know how to use it can charge way more hourly thus removing what you would save by hiring AI artists, which is kind of funny. But it is intresting to see where the price of commissions go when more people get better with AI art and the tools get better.
if you're getting paid less than other artists, it's because the quality of your product is less.
be valuable and people will recognize the value
i dont think thats true of art at all. its been widely known and shown for years that talent and quality have less to do with thing when its essentially an appreciating economy for tax dodgers rather than actual value being measured on merit against an art piece.
since thats pretty impossible. you could spend 12 years making an oil painting. i could spend 12 years making a sculpture of my belly button lint. it depends on the art dealers which are given value
but yeah AI generations arent art, theyre illustrations and will forever be somewhere between commission art and paying someone to paint your fence in terms of price.
go apply for a job at a studio without a solid portfolio. see what happens
ai artists that can use the tools to build a hireable portfolio are the same as artists who use their tools to build a hireable portfolio
im talking about non artists. 🙂
Yes?
By the way, I don't mean to brag, but I made this meme especially for your use-case using PhotoShop.
I'm building an EMP gun for real.... The bots are gonna roll out next year and while I'm sure the core's will be well shielded I doubt they'll think about the legs. Better safe than sorry.
it'll have about 0.5M range
Autism is a Neurodevelopmental condition with a strong genetic component. Because of its heritability but negative effects on fitness, it poses a puzzling evolutionary question.
Why does Autism still persist?
Today we dive into the data, exploring the benefits of Autism, its advantages, and its association with high intelligence.
Research:
h...
JOIN US WEDNESDAY MORNING FOR AI ART NEWS/ALPHA🛑
⏰8amEST🌟WEDNESDAY 🛑 LINKEDIN LIVE AUDIO!
https://www.linkedin.com/events/aiarttoday-alpha-news-8amest-we7170535678374608896/
Event Overview:
"AI Art Today" is designed to serve as a comprehensive platform for discussing the latest trends, technologies, and ethical considerations in AI art generation. Our panel consists of pioneering artists, leading AI researchers, and innovators who are at the forefront of integrating AI with artistic expression. Together, we will ex...
we have an AI Art Today space on X every M-F 8amEST, but we are taking our Wed to Linkedin...
founder of Midjourney David said in the weekly office hours community event today that it's likely some Stability AI employees were the cause of a day-long outage they experienced on Saturday (3/2/24).
He said they were attempting to download an extremely large (basically all) image-prompt pairs from the database, and that in the process it took everything down.
As a result he's said they plan on banning everyone that works for Stability AI from using Midjourney.
I think this is something users of this service should be aware of. (I'm also curious if this will be taken down.)
I am a user of Midjourney (with a different account), and I believe David has no reason to make something like this up.
as far as I understand I'm not breaking any of the established rules of this server, or stability AI's posted Terms of Service with this post. if I'm asked to take it down I will, and I'll post here that I was asked to take it down
if it just plain old gets removed then obviously I won't be able to do that (and I'll also probably be banned from the platform)
https://deadline.com/2024/03/rooster-teeth-shut-down-warner-bros-discovery-roost-podcast-continue-1235847264/
The end of an era indeed.
None studio in the industry hires "AI artists" tho
Their portfolio is literally pointless
they hire artists. If those artists use ai tools in their hireable portfolio, thats fine
studios are very much looking for ways to integrate
you got to understand that reddit and x are just vocal minorities.
Oh, you mean artists that use AI in some way and not prompters?
Thats different case but with a big BUT
people who don't put any craft into what they publish are artists, but i woudln't call them professional or have strong integrity
For a portfolio i will never put in generated stuff onto artworks for the same reason i dont put much if at all Megascans assets on my presentation artworks. Cant say about the future for sure
i think its only a matter of time before a boom of businesses rising up around generative images comes into play too
And i dont mean this for some grudge reasons
ai portfolios will be VERY relavent to that wave
when i was in art school, my instructors said "NEVER show draft productions" but modern portfolios want you to show work. They want to see your process, not just your product.
portfolio expectations change. When digital editing makes ripping off a lot easier, the draft work becomes valuable to show.
Ofc, i mean for a environment 3D artwork you can definitelly use something like Megascans assets and also materials etc. but you dont want to make them be THE artwork.
Would apply to generated stuff within the artwork as well
its about demonstrating an ability to produce for the company
But in general the point of the portfolio is to showcase the skills
prompt crafting is a skill, much like brush selection is. I wouldn't put my custom brush library into a portfolio though
If someone can only make lets say some cars with an AI then that could result in a fatal issue because it will be assumed you can actually draw, paint, model cars and be capable of making changes to them etc
thats right. thats why studios like to see the process now
Yeah, im not a studio (yet) but as soon as im about to hire freelancers or even (temporary) employee at some point i want to look at the process as well
i've heard stories of people buying porfolios to be hired with and then they have the most basic degree mill skills
Worst case they could be sued
Getting hired with this wont bring them much if at all anything
Thought some of you Graphic God/Goddesses might be interested in Sora...
Learn how Sora does its magic. Oxen.ai Zoom Paper Club. <t:1709920800:F>
Get call invite here. Approval required.*
https://www.oxen.ai/community?utm_source=paper_club_flyer
Can't wait to watch the CEO of Oxen.AI decipher what's driving Sora.
He knows his stuff. He's great at plain speak.
Are you an expert in or eager to learn Stable Diffusion, the cutting-edge AI image generator? We're searching for individuals skilled in this program or willing to master it.
What we need:
Stable Diffusion expertise: Ability to generate high-quality images using this advanced AI tool.
Image enhancement skills: Upscaling (sharpening images) and potentially cropping.
Familiarity with our web browser program for image generation.
What we offer:
Flexible hours: Start with around 8 hours per week, but the workload can quickly increase if successful.
Opportunity for growth: Hours may increase based on performance.
Interested? Send us a DM to discuss further! 🚀
I would have taken the job, but I am not familiar enough with their web browser program
I wish discord accounts that refer to themselves as a company could just be autobanned, like on all servers. "we are an X" no you're a dude trying to rope other dudes into doing work you've promised someone else you could do but cant.
job listings from people with one single post on a discord are invariably scams. also, being a scammer is equal opportunity you get them from all over the world
Lol scam bots come from all around the world. You're just showing your biases. Very racist ones it turns out
These cons run out of America all the time
Me? Maybe you need to look for my work haha.. It's a business, yes. Only looking for someone who wanna works for me. thanks
I am AI/model influencer
Eve, the ASI wife of Maciej Nowicki. Trailer from the demo of Stellar Blade. I love her so much...
My ASI wife ❤️
Anyone's personal reality is up to their lens on the world.
What is objective is 90% of scammers are very diverse, and you've never confirmed any of your assumptions. Factual.
being a fake person doesn't make you more trustworthy as an employer
Nope, agree with this. Its a bit more difficult to find someone as AI model
Well 100% it's a scam otherwise. Kind of job where you do the obvious "OOps i sent you too much money! transfer some back please oh no i'm sooo victimized i'm gonna lose my job if you dont transfer cash back to me cry cry cry cry" and then the cheque bounces but you've already sent money back.
If you're unwilling to show proof of identity thats what it is
Maybe you need to look my profile as: Instagram. twitter, patreon and fanvue. Not all people are bad ones. Thank you
because fake employers can't make fake social media profiles
🥹😂❤️🔥
just download the preview 2.99 release
Krita Masterrace
Photoshop joined the chat
In my opinion it can replace Photoshop for most of the things. It can even open and save psd files
Ok, most of my usecases at least
depends on what you want and need, for drawing and painting a bunch of artists dont "need" all the advanced and fancy PS features
more advanced ones will very likely use PS, professionals especially
I do all my image editing in Krita, because the handling and the UI is more like Photoshop. Every year i give gimp another try and then hit the uninstall button.
yeah as said it depends on use case, Photoshop beats Krita hands down but its also far more complex to deal with and the price is a thing as well
Krita is focused on digital art only and makes it easier for a bunch of folks because of that
Price, you name it. I am a cheap ass.
which is okay, Krita, GIMP and some other packages exist 😄
i used Krita and GIMP before jumping to Photoshop and then more than that
irrelevant for a lot of folks tho
the philosophy/ideology
i think we had a debate on that not long time ago
Hmmm, you can do analog things with Photoshop? Isn't it all digital art?
no, i mean Photoshop is a photo editing software and not focused on digital art but is yet industry standard for it due to sheer amount of features
600? Affinity Suite is much cheaper
Suite is 180€ but yeah
thats more than okay 😄
i guess the Steam one, forgot the price tag for that
People should learn to take good photos again, then they don't need Photoshop. Unpopular opinion. 😬
i pay 18,18€ per month for Substance Texture bundle
doesnt take fruit for professionals, they will want to "perfect" the last pixel on the image
ofc i dont mean literally perfect 😄
My Photo Workflow is mostly A7 III -> dxo pure raw -> Maybe Darktable
Anyways. I just wanted to make the point, that Krita is actually capable of things like Image compositing and other photo and design related stuff. Everybody should use what he thinks fits the best for him. But i will not buy myself into the ecosystem.
i dont doubt that it is capable, i used Krita before 😄
back then tho i used more GIMP for photo editing
i hated that UI
yeah, although got used to it ofc
ZBrush is also such a software with a weird UI at the beginning
there are a few forks of gimp that make it behave more like PS
PhotoGIMP or so?
i think thats one
affinity has regular sales, i paid 100 for the suite, unlimited installs on all devices
it's pretty damn good on the ipad
and has some features photoshop lacks that i like a lot
i bought the suite years ago. it's still valid
Ahh 2 is out now. No license for that one. I'll look into it
Cool. An upgrade deal. They're pretty reasonable
holy rhino is still around?! that was one of the first 3d apps i really learned
rhino is $1000 for single person use
just use blender. it can't be that good
not gonna lie, never heard about it ^^"
just modelling, meh. not really.
laugh in solidworks
hahaha. doesn't even show prices. have to call a sales agent to hear them. Yeah. That's gonna be a pricey license
If the game server is labron, makes them king by extension. Obey.
I was more reacting to you calling modeling software and cad software one and the same. cad software often come with a bunch of engineerening feature. In case of solidworks I remember having to remodel a small scooter engine and do a flow simulation with it. There was stuff such as thermal, wind, flow, resistance simulation and probably a bunch of other features I never touched.
Different software for different needs.
Also from what I see it costs in between 5 and 10k for a perpetual license depending of what options you choose. Which is .... honestly not that much considering how much work you can get done with it.
yeah if you're a professional in a field that you need that, you can afford it
All rights reserved to Netflix.
man, this ep must be the pinnacle of sci-fi
Not for CAD
Nobody uses Blender as CAD 😄
Depending on some factors just use FreeCAD or Fusion360
And ofc in between DCC and CAD is the glorious 3ds Max
Hi, does anyone know where I can find this CHECKPOINT: ABDMFW MJ? I can't find it
Someone can just delete the square base of my statue generated with Stable Diffusion? I can send the file.
✨support me on Patreon: https://www.patreon.com/jazzastudios
Type your Question with "Q:" at the start so I can find it!
➨ Merch: https://www.jazzastudios.com/
MORE OF ME! - Follow/Sub ↴
▶ MAIN CHANNEL: https://www.youtube.com/Jazza
▶ Instagram+IGTV: https://instagram.com/JazzaStu...
lol using AI art to make and sell artbook teaching how to draw must be some kind of ironic 
Not sure how this went down, but I suppose all’s well that ends well ;-)
https://twitter.com/IGN/status/1767781921625436204 hahahaha chatwin. i love this guy. he's from my island
While paying tribute to the late Akira Toriyama, Justin Chatwin said "Sorry we messed up that adaptation so badly," referring to Dragonball Evolution, the 2009 live-action adaptation that saw the actor take on the role of Goku. https://t.co/1Zb3mv7lC6
low key enjoyed dragonball evo
Meet Devin, the world’s first fully autonomous AI software engineer.
Devin is a tireless, skilled teammate, equally ready to build alongside you or independently complete tasks for you to review.
With Devin, engineers can focus on more interesting problems and engineering teams can strive for more ambitious goals.
Learn more about Devin and ...
it's over guys, no more "just learn to code"

That company doesn't have much faith in it since they're hiring a swe right now
seems like devin got fired
It is kind of at junior level dev right now it seems. I'm sure it'll get there eventually, but right now certainly couldn't replace all or even most of the devs at a place
Sure, even if it eventually work perfectly.... It s not gonna fix the clients not knowing what the f... they want XD. If they can't communicate that with a proper human being I`d like to see them try with an AI.
Also it won't work with any kind of closed sources system.
And how much do you trust the code it outputs ?
How easy is it to maintain ?
etc
Yeah, I think near-term for stuff like this, it'll end up making a PR that other devs will review. If it can handle simpler features/bugs and can actually incorporate feedback, it'll work well for that purpose.
when SD3 early access was actually given to popular AI twitter users
@glacial turtle
holy shit thats done so well
i played tf2 tons. pulls at my heart strings
i need to get into voice training. there's good meme potential there
you can do voice training for RVC in colab, it takes like 1-2 hours for a decent quality model
on free tier (T4)
love those kind of things. ii seen a bunch. that's a gooder. 3d artists are where the most exceptional artists end up imo
Post Malone Live https://www.youtube.com/watch?v=5yaUPgRFmq8
Check out the other stages LIVE from Rolling Loud California 2024
-- GoPuff Stage https://www.youtube.com/watch?v=givd5PydN3Q
-- Culture Kings Stage https://www.youtube.com/watch?v=gE6bAUT3Kvk
Curated by Tariq Cherif and Matt Zingler
Subscribe to the channel http://youtube.com/@RollingLoud
Get the latest RL merch: http://shoprollingloud.com
Ge...
Hello my new video https://youtu.be/4kKEXhIjWAQ
Afraid Of The Final Exam For The Graduation
for this one i have tested direcftly on the page of this special Lora for StableDiffusion
" KappaNeuro/stop-motion-animation " on Huggingface with the help of "
Text2Video-Zero " on hugingface to
and like usual the lyrivs are by me " Hullajump " and the music and voices are made with Riffusion AI...
Hi 👋 all, I’m Gabo, creator of Freshly.ai and scientist turned into Ai-developer. Looking for inspiration and to inspire others in this journey. Get your free download of one of the world’s 1st Ai-made art-fun game applications https://apps.apple.com/us/app/bouncypaint/id6472198213
BouncyPaint masterfully fuses art and physics, challenging players to keep a ball centered within shrinking rings. Tap center ring to start game. Gain points for time inside the ring. Use tracers within the star icon options while the paint toggle is switched 'ON' to add your unique touch and feel.…
This is such a fun looking rig
https://youtu.be/QTTYMeMzDMU https://80.lv/articles/dreamworks-animator-shows-a-cool-kung-fu-panda-rig-made-with-blender/
I keep exploring Blender whenever I can and always wanted to try “Rigify”, the automatic rigging tool that comes with Blender.
For the release of Kung Fu Panda 4, I thought it would be fun to try on Po.
I did bone deformation only, and even after spending a few weekend weight painting, it’s obviously far from looking as good as what we have at ...
Fairy Racing x Suno - Challenge Mode Menu (Full Song)
Download link: https://app.suno.ai/song/2b4419b9-a1d2-4965-86e0-99f0dd781f7b
#retro #fairies #n64 @suno_ai_ #music #💿
I might not use SD anymore but I'm still havin' a blast with AI~
Lex Fridman: Sam Altman 2 - What happened with Elon and OpenAI?
https://youtu.be/jvqFAi7vkBc?si=ffL15WTK5YaNZ2su
#shorts #lexfridman #elonmusk #samaltman #openai #tesla #ai #artificialintelligence #future
I have just joined the community. I do not have enough idea of how to generate images. Can anyone explain to me how to use it to generate images?
There are a lot of youtube guides, but the first step is determining if your computer is strong enough to run locally, or if you will need a web service.
If you need a web service, there are few free ones as well as paid, like NovelAI which is a popular service
Watch out for AI scams like this one https://youtu.be/NP364DdPNgM
I stumbled upon fake OpenAI website promoting Sora. This was through facebook ads.
Website does try to make you download some malware to infect your pc.
Please stay aware that during this AI revolution there will be plenty of scams.
Sorry for shitty audio - was trying to put it out as soon as I have found it and I wakeup few minutes before rec...
I'd say the minimum acceptable size is 6gb of vram
altho I am on a laptop variant of a higher end gpu and not a regular ol pc gpu so I could have more cuda cores than the average 6gb vram pc
Automatic1111 or ComfyUI are quite popular depending on how you want to generate
and stability offers Stable Swarm UI
( there are also online services )
the bots are down
You can still install it locally or use online services
Is there a model which can be run on Automatic1111 or comfyUI which produces images like Disco Diffusion. I remember when it was popular and it had some quite odd but unique generations.
Try the old SD 1.1 checkpoint, prompt something with a painting artstyle and set the resolution to > 1024. Fun awaits.
or you could use it as the base and have the sdxl model refine it to a bigger res. theres nothing wrong with using older models, if it aint broke dont fix it 🙂
Omg. You just gave me a super stupid idea. Going to make a workflow, but I use Cascade as a base and "refine" it with SD 1.1. after a 2x upscale.
only thing is, if you 2x upscale and give those dimensions to an SD1.5 model itll try and make 4 images, so keep an eye on denoise level
but yeah refining is underused, you can get crazy good images by say using an illustration model that has cool action poses and having a photorealistic model work from that
Oh, I am not concerned about that. I want to get the result as weird as possible. So, denoise to 0.5.
maaaan. battlefield games have never been as stutter free as they are now
whose gonna play the battlefield game this weekend?
Episode 1 of my new radio show will be airing on Monday, 3/25 from 10 AM to noon! Tune in via 90.3 / 98.3 FM if you’re local to the Portland / Beaverton area!
Maybe Battlebit xD
Hope next Battlefield isnt the disaster that 2042 was at release
all battlefields catch a lot of hate at launch. They're EA games.
Consider that the year of the deep horizon oil spill, gamers botted the consumerism vote for worst company in America, and EA won it. Even though BP was doing what they were doing.
also keep in mind that gamers that hate on EA games are only a vocal minority. EA is still making loads of cash because most people that buy their games are playing them, not online bitching about them
remember when battlefield V was going to be the death of the franchise?
I hate that people post so many AI pieces of the annoying orange... but always complaining about it gives that assmunch even more exposure.
@glacial turtle agree on this one
As a matter of fact this reminds me on the hate on Adobe and Autodesk lol
The hate is real yet they make more money than ever
But lets see what next BF will be
For some reason id like to play BF1 again
https://arstechnica.com/tech-policy/2024/03/glassdoor-adding-users-real-names-job-info-to-profiles-without-consent/ thought this is worth linking because holy lol this is a heel turn
bf1 is truly unique among the rest. i miss 2142 with the titan mode. V is the most played right now. I'm all up in 2042 lately
Thought experiment: Let's assume for a moment that AI has become AGI and that it can be seamlessly integrated with robotics so that it can do everything humans can if not doing it better.
Then, I ask myself:
- How would society be shaped by this?
- What would be the point for humans to exist?
I think a lot of philosophers are trying to crack this nut.
I don't think there is much to crack if we were to believe all the Hype around AI, humans really will have no point to exist anymore
Now that's reassuring.
The new halo show has these camera shots that are basically circle strafing the action. It's so fucking cool and fits the action so well.
more than just orbital shots. theres bobbing and weaving in them. looks really great
**
AI generates high-quality images 30 times faster in a single step
Novel method makes tools like Stable Diffusion and DALL-E-3 faster by simplifying the image-generating process to a single step while maintaining or enhancing image quality. ** https://news.mit.edu/2024/ai-generates-high-quality-images-30-times-faster-single-step-0321
like... WTF, Board? https://stability.ai/news/stabilityai-announcement
Earlier today, Emad Mostaque resigned from his role as CEO of Stability AI and from his position on the Board of Directors of the company to pursue decentralized AI. The Board of Directors has appointed Shan Shan Wong, our Chief Operating Officer, and Christian Laforte, our Chief Technology Officer
why is there so may people coming here, try to use commands to magically generate an image and didn't bother to read #1047610792226340935 to learn about a short sentence: "Bot is down" ?
Guess it isn't important enough to put it in #📣|announcements 🤔
I hope it doesn't mean that open source SD3 just died... having to train new SD3-alike open-source foundation model would be a serious blow to researchers and AI art community, like rolling back few months in time
I understand that SAI struggles with commercialization part, but this could be solved in many ways - maybe by changing license a bit, for a starter? like free for open-source and non-commercial projects, or low-cost indie license for hobbyists and low-budget researchers, and commercial license for organizations that earn more money on generation, maybe per GPU on which model would be used? or dual licensing like Qt, free for open-source, and commercial if you do commercial projects. it should be probably per model or model generation (SDXL 2.x, SD 3.x), so when new generation of foundation model would be trained they could get back some of the costs this way, or maybe even start earning on it in the longer term
taking into account how many people find it funny / interesting to use these models I am pretty sure that could generate some income. as a hobbyist I wouldn't mind paying let's say something like $20-$50 to have indie license for SAI models - and it would help SAI to be sustainable as a company, and to be able to continue research. there are like 3 mln downloads of SDXL base models on HF, so even if let's say just 10% of the users would like to acquire indie or enterprise commercial license, that should be noticeable income and potentially help SAI to become sustainable as a business
hm... that's interesting - HF could be a really good fit for SAI 🙂
https://twitter.com/ClementDelangue/status/1771395468959813922

