If you like Funky songs here's one ! https://www.youtube.com/watch?v=6D8nQNKTv50
#🌶｜off-topic
kind spade
fast oak
this SSD is toast
zealous cipher
hello, Hugging face released their own GPT, any idea on.. how do they make money..? ive heard that CHAT GPT cost 400K a day. how hugging face have the money to run a chat gpt like on their own ?
sturdy nacelle
from now on I'll use chatgpt to speak to people on discord
holy p they interpret a joke too seriously
would this keep discord folks happy
hot otter
hello, Hugging face released their own GPT, any idea on.. how do they make money...
- https://techcrunch.com/2021/03/11/hugging-face-raises-40-million-for-its-natural-language-processing-library
- https://www.forbes.com/sites/kenrickcai/2022/05/09/the-2-billion-emoji-hugging-face-wants-to-be-launchpad-for-a-machine-learning-revolution/
- https://huggingface.co/pricing
etc etc... That's how.
zealous cipher
- <https://techcrunch.com/2021/03/11/hugging-face-raises-40-million-for-its-natu...
thanks !
opaque dawn
Crazy, now Hugging face has their own chat model :/
flat pumice
If I want to change my mind a little, like RPGs and space, and narration :
A/ Outer worlds
B/ Outer wilds
?
arctic swift
We're building a platform to help freelancers promote their podcasts, videos, & host real time community events.
https://twitter.com/MentalWealthDAO/status/1650919330341593093?s=20
You can donate ☕️ ^ through @gitcoin
Donations are amplified so even $1 could mean the world for the project and others 🌎🙏🏾📿
south thicket
cobalt birch
Anyone interested in joining an AI-art server, focusing on tarot, mystic and occult generations?
🔮
Dm me
glossy raven
Helloo
opaque dawn
trim dragon
Trailer for my new album, releasing May 7th!
jagged path
pretty cool to see what someones doing with AI
not sure if all the pics are ai gen'ed or just fanarts but these are some pretty entertaining vids, learned more warhammer 40k lore than i wanted to
pearl wasp
https://youtu.be/7D-VBrRdfVE
no github link?
atomic tree
pretty cool to see what someones doing with AI
i think it's a really bad look for the tech when the first stuff that gets popular with it is plagiarism of true talent's likenesses.
They could've easily made just as high quality videos using their own voice and training that so they didn't have to read lines, but they wouldn't have caught the views they could get from ripping off attenburough. Click bait is why they ripped off attenburough. No other reason. I think it's a bad look for us all and not "pretty cool". Sucks to see people doing that.
Another key part of attenburoughs work is he is always on site. Just feels so immoral to rip him off.
jagged path
i think it's a really bad look for the tech when the first stuff that gets popul...
I see your point and it's a good one about plagiarism but at the same time, I definitely would not have watched the vids that I did if not for the same format of Attenborough narrating things that I'm used to.
lofty yarrow
thats what this guy is doing with his life?
winged pine
ornate garnet
Does it probably mean that Twitter will sponsor stable-lm?
last stone
Does it probably mean that Twitter will sponsor stable-lm?
0.o no, Emad is saying in that screenshot that he's taking a break from posting on twitter. Which I think only lasted a day anyway
worn pine
I am haha. This video is entertaining in its own right though 
thats a tough question to answer and I would say #1 is it lies mostly in if you find it interesting, its never worth doing if you dont get some enjoyment out of it. I would try some easier programming intros out there on the web (not sure what is best at this point) and see how you feel about it. I find programming tends to feel like puzzle solving mixed with creative expression, good bit of joy from both ends.
#2 is that the field is certainly changing pretty quickly with all the advancements in LLMs. I honestly don't know what programming will look like in 1-2 years from now. That all said, becoming more technically literate and understanding how the software you work with operates is never a bad thing to get better at 🙂
definitely looks like its going for those "realistic imperfections" with the overblown camera flash and random stuff in the background. Good for the realism on a mediocre camera look
boreal marlin
Hey everyone!
I’ve personally found it difficult to ever finish my to-do lists. There is always more and more to do. During a recent hackathon, we began working on a to-do list app that we thought might be much more helpful. A to-do list where the AI actually does the task or helps you complete itself!
Imagine a to-do list where you can delegate your tasks to autonomous AI personal assistants who will help you get it done… or at the very least, plan, coach and motivate you to get it done. We’re looking for some early alpha testers and if this is of any interest, would love to chat more and see how we can help.
sleek flower
fast oak
Hi! Was wondering if anyone knew of high quality image captioning models which are able to generate detailed descriptions? Most of what i've seen is about 20 tokens max. I've also been thinkging about combining these with GPT to generate questions + use question-answering models, but it feels convoluted
olive snow
atomic temple
Hi! Was wondering if anyone knew of high quality image captioning models which a...
u can use BLIP from salesforce
it was like $3 for a 1000 images when i last used it
fast oak
I have a bid for a used like new glacier DS4 for about 29 euros It's a good price right?
meager tree
Big release here using Stable Diffusion on the backend. RTs much appreciated!
https://twitter.com/buildcities/status/1653483955243614208?s=20
Twitter
torn cove
drowsy moth
Funny game night highlights from a server I’m in. One of the guys actually knows how to edit videos
junior meadow
https://youtu.be/V-lspfpHoKY
any model suggestions that can imitate the stretched lights/speed light seen in the night, like in this one?
flat pumice
https://www.youtube.com/@wallcraft-video
interesting youtube channel, where the guy simulates life and comments on it
fast oak
u can use BLIP from salesforce
It's open source actually - but it still seems to generate pretty short captions (<20 tokens).
olive snow
any model suggestions that can imitate the stretched lights/speed light seen in ...
that specifically is not ai generated
it is a video i picked up
ai video generation is pretty primitive atm
opaque dawn
Don't think MJ 5.1 is too different: https://youtu.be/8mn_mGqkKpk
amber vale
that specifically is not ai generated
Yes, no doubt that video was reality based. it's just that some models are specific for different kind of visuals. i wonder if there's any model that's good with night lights, especially night/low light photography. (Btw, Deforum extension is a lot capable for video purpose)
crimson bone
Drawing, Designing and other Creative Sessions on: https://kick.com/3esign
zinc nest
Hey guys 🙂 we plan an twitter space about AI ethic - https://twitter.com/i/spaces/1ynKOaOABbkJR . Feel free to join. 🙂
lusty sluice
hybrid zealot
what are your thoughts about leonardo ai?
it seems like an easy version of stable idffusion
opaque dawn
AutoGPT, but FREE: https://youtu.be/KlOB-3sY5gg
lusty sluice
first time in the community : https://www.youtube.com/watch?v=343I11mhnXs
i already have here 2 videos
still working amazing up to date
Really good
paper nimbus
Hello, if anyone can help me with these, we can discuss and amount and happy to pay.
Sure. Here’s are a bunch of things I need to do, lmk if you can help me with any of these.
- I have a fitness model pictures in a folder, and I have a lot of it. Highlighting her face, shoulders, legs etc…
Now, I want to create images with different poses with it using stable diffusion, so that the model don’t have to take pictures anymore. Also by using control net with SD, I want to generate images with specific poses that the model hasn’t poses for before.
- Now I have the models image, I want to change the background of the model. So 1 model with the same pose, different types of background. The background shouldn’t be much different from one another, ie they should be more or less a little variation of each other.
opaque dawn
10 Prompts tested across different subjects:
icy quartz
https://www.youtube.com/watch?v=m8W4l-peEBk
I read "at grandmother level" in title and then grandmaster in thumbnail lmaoooo
fast oak
If you didn't care much about speed. Could you run stable diffusion on M1 or M2 with 64gb of ram to build higher quality images opposed to 8Gb Nvidia gpu?
lusty sluice
I read "at grandmother level" in title and then grandmaster in thumbnail lmaoooo
haha :d
rough arch
https://www.flyingdog.de/sd/ this is so much better than any free plugin or adobes firefly
leaden kelp
crimson bone
Join me in live art making sessions on https://twitch.tv/3esign | https://kick.com/3esign | https://youtube.com/@3esign or check out all my streaming links on https://linktr.ee/3treams | I am a PhD Architect, Poturak Semir, teacher of art and design - follow up for my upcoming art school < 3
topaz tiger
hybrid zealot
why does midjourney offer free trial now , do they get threatened by other free image ai generator?
shut crow
My ssd has 91% health and i built my pc beginning this year
Wow what Brand is it ?
hallow nymph
why does midjourney offer free trial now , do they get threatened by other free ...
midjourney always had a free trial??
hybrid zealot
midjourney always had a free trial??
not always.. maybe you didn't know they remove the free trial many weeks ago
tidal whale
Are these specs good for running Stable Diffusion? I’m new to all of this
south thicket
Are these specs good for running Stable Diffusion? I’m new to all of this
Processor is overkill. Gpu should be excellent.
16gb of memory is a bit lacking. Would be worth doubling
tidal whale
Thank you for your feedback!
zinc storm
Wow what Brand is it ?
Samsung
Remember
There were a firmware issue on 990pro series
fast oak
Are these specs good for running Stable Diffusion? I’m new to all of this
Dang A10G 
flat cove
https://www.youtube.com/watch?v=z-Ng5ZvrDm4
people want this future, how should we build this
plush gorge
hi guys
i have problem
why my generates goes this?
in realtime render all looking good but in final...
agile river
you probably need to use a VAE
icy quartz
why my generates goes this?
did you get it fixed?
plush gorge
you probably need to use a VAE
This model does not need vae
icy quartz
which model
plush gorge
awawa...
I'm forgot
xd
But when I'm go back from sleep I'm dm you name of model
Okay?
icy quartz
ok
hybrid zealot
anyone have any idea how this works? https://www.getit.ai/gpt-plugins
i dont understand basic coding as i have no knowledge about it, it only gives code when i click the plug in
junior crow
anyone have any idea how this works? https://www.getit.ai/gpt-plugins
I'm so glad I learned Ruby! I'd probably go insane if I had to do anything in Python.
junior gale
car
harsh osprey
Hello guys, i've started to use stable diffusion like 3 month ago, and managed to create this music video, i'm a musician from Kiyv, so hope you give me your opinion about this work, here , or better on youtube😌
vocal raven
pretty good, I prefer it to most of the videos that go full out on using SD on everything in every shot
opaque dawn
Check this out! 😮 https://youtu.be/bSYzDlEnEGE
flat pumice
Outer wilds was so good... long time I didn't have such a good time on a game
tropic ivy
do anyone knows how deforum have been built?
and how can i make extensions relying on sd?
things like temporalkit and discodiffusion
anyone have anyidea?
icy quartz
that seeems like a really non-standard model
you need to use the specific VAE it comes with
plush heath
Can I ask for you guys for a opinion...does this guy seem suspicious? Would you continue doing commissions for him?
tidal whale
What is a prompt I can use to make a person stand far in the distance?
Thank you!
Also one where the person does not look into the camera!
plush gorge
that seeems like a really non-standard model
this model broke after minor update)
south thicket
Can I ask for you guys for a opinion...does this guy seem suspicious? Would you ...
Nah don't trust him
Super sketchy.
I'm guessing he's going to issue a chargeback of cancel the payment somehow because you 'failed to deliver what he asked for' or some silly nonsense.
plush heath
I'm guessing he's going to issue a chargeback of cancel the payment somehow beca...
Well he cash app me so idk if he can refund on that. But, I started catching on when he was trying to make it "my fault" by asking "if I only do censored"? He told me to censor the parts so idk why he's asking that. So I'm not gonna work with him anymore.
I was so confused as why he's happy about it....
Little does he know I have them in my rubbish bin but eek
opaque dawn
Bard is pretty good lol! https://youtu.be/2d1hlqlUR6Q
weary jasper
modest kindle
Hey fellow Diffusers! 😎 You know that feeling when your Stable Diffusion spits out so many dank meme upscales and rad Greg Rutkowski art pieces that you feel like you're lost in the sauce? 😵 And you're sitting there, scratching your head, wondering which masterpiece is the true Doge among Shiba Inus? Yeah, we've all been there.
So, I went ahead and built a little something to help us all out. Introducing TourneySnaps (www.tourneysnaps.com) - it's like a royal rumble for your images, a battle royale where only the dankest hires realistic meme or the most breathtaking waifu can emerge as the true Jar Jar Binks. 🫙 👑
Now, you might be wondering, 'Am I gonna have to upload my precious astronaut on a horse pics to some random server?' Absolutely not! TourneySnaps is a simple Javascript app that works entirely within your browser. Your images don't go anywhere - they stay on your computer, safe and sound. No servers, no third parties, nada. 🙅♀️🚫
Feel free to check it out and give it a spin. And if you've got any thoughts or ideas on how to make it more lit 🔥, I'm all ears.
west void
Anyone know any good opensource thing to generate simple no vocal music? Around 1 minute
maiden cove
I am analyzing ChatGPT traffic patterns and discussing ChatGPT's effect on Stack Overflow and GitHub.
https://www.youtube.com/watch?v=XH63jCFUSwQ&list=PL-ciccwnzEopTel0io2qEYPaHxRVLK9FX&index=2
unkempt scroll
copper folio
Hey there 🤓 if you have a moment, could you give me a hand? 🙏🏽
This is my first attempt to release a product online!
worn plover
https://www.youtube.com/watch?v=75m4X3EjZd4 Grimes used for Vtuber female rapper project
soft flare
brain hero in a zombie world
tidal vigilBOT
soft pine
opaque dawn
super cool: https://youtu.be/HxPquLttdDo
soft pine
ember dome
Hello, i dont knwo if this is the right chat but im posting it, feel free to delete if im in the wrong chat for such things. im getting into stable Diffusion and planing to work with Sadtalker, thin plate spine motion model,....
I found that my AMD GPU is not the right choice for it and i got told i should choose the INNO3D GeForce RTX 3060 TWIN X2. Thats like 340€ here in Germany and its my absolute max investment i can do.
What do you guys think?
is this the best choice for this Price?
Should i choose exacty this one or one of the many other 3060s?
I got told that i definitly should choose the 12gb version.
I could take some advise. Thank you
light galleon
I would definitely stick to the nvidia offerings. I use a 3060 12GB and it works fine. The more vram the better.
The other aftermarket gimmicks (OCs and coolers) don't really matter all that much.
cloud osprey
feeling so so hungry right now
narrow willow
Hello there, anyone having knowledge about unreal engine? Ready to pay for your time, just need some information
vagrant rover
(loud) new musicLM by google with the prompt "8 bit, rap beat, slow tempo, smooth and groovy, ambient, solo", doesnt follow the prompt when getting complex at all but its like early text2img prompting, u didnt really hope for the actual subject, just for that area/field in the neural network (if u understand correctly)
deft vigil
Hi what YouTube video dowloadeder do u recommend that doesn’t mess up the quality of the video and allows you to download the highest quality for free?
opaque dawn
INSANE! https://youtu.be/0qh0CzRNNj4
dark meteor
BITCONNEEEEEEEEEECT
dark meteor
(loud) new musicLM by google with the prompt "8 bit, rap beat, slow tempo, smoot...
my shot at it: "a lively track for like when entering a town in a mmorpg game, very happy tune with flutes and harps"
supple shuttle
my shot at it: "a lively track for like when entering a town in a mmorpg game, v...
sounds like when you meet the annoying party memeber you cant kill off.
vagrant rover
my shot at it: "a lively track for like when entering a town in a mmorpg game, v...
nice rlly accurate lol fun giving a prompt to text2img and the text2music, works rlly well together, i found its best to take text2img prompts and give it to the music one rather than the other way around but both are fun
vagrant rover
my shot at it: "a lively track for like when entering a town in a mmorpg game, v...
"A dubstep track with aggressive bass drops that breaks into calm downtempo sections, complex drum patterns, and chopped-up vocals, characteristic of the early 2010s"
dark meteor
I also noticed it performs quite well with dubstep lol! 🎉 that's 🔥
vagrant rover
had gpt make a bunch of artist related prompts bc it doesnt support em :/ that was skrillex with some changes
dark meteor
Discord really did us dirty man
heavy lava
is this website joke https://text2icon.app/
i think it is joke
can some body test it
craggy hazel
looking for a tool which helps me make novels. like sudowrites new feature called "story engine", but cheaper.
shut crow
looking for a tool which helps me make novels. like sudowrites new feature calle...
Novelai maybe
craggy hazel
Novelai maybe
ewww
unuseable trash like a trashcan filled with condoms
nah sorry. novelai is nothing, like gpt4 vanilla for making good LONG stories
shut crow
Yea then im out of ideas, would have more ideas if you search tools for image generation
craggy hazel
i had enough image generation. i saw everything.
shut crow
There are a lot of llms maybe something can be helpfull
But i see that a already set up Website would be more convenient
lapis lodge
made a funny wes anderson themed trailer using harry potter characters
crimson marsh
How do I remove myself from challengers you keep pinging my discord at 6am.
maiden cove
I compared 17 different ChatGPT alternatives, free chat apps, Bring Your Own Key, prompt templates, and nerd out a bit comparing tech stacks at the end. Would you mind sharing your thoughts? Also, I would appreciate any corrections or suggestions you may have for the spreadsheet comparison table.
https://youtube.com/watch?v=QFG8QOdS148&list=PL-ciccwnzEoobqguX_8OaufwCuuJTtoZ4&index=1
deft vigil
What paid upscaler and image enhancer is currently the best software or website paid or not but most likely best is paid so please recommend me some that are actually good
Ty
naive stirrup
opaque dawn
soft ingot
agile river
What paid upscaler and image enhancer is currently the best software or website ...
Topaz Gigapixel or SD Upscale Script, in my experience are the best.
opaque dawn
Took some Stable Diffusion images and edited those over generative fill, pretty neat! https://youtu.be/V6XuEAuxFZk
cursive flame
How can I see blurred images?
long obsidian
atomic tree
How can I see blurred images?
bokeh invokes crazy blur photography
meager depot
Should I upgrade my CPU? I just got a 3090 and I'm still running i5-6600
south thicket
Should I upgrade my CPU? I just got a 3090 and I'm still running i5-6600
Depending on how much upgradability you want, I'd recommend one of the newer asrock z790 boards. They have support for 12th, 13th and rumored 14th gen.
Though if you're just looking to get the most out of your card, a 12600k and a cheap z690 motherboard with high-speed ddr4 will do the trick.
meager depot
Depending on how much upgradability you want, I'd recommend one of the newer asr...
I try not to upgrade much so next upgrade might in the next 6-7 years. I think I'll look into your second suggestion thanks 😊
split yew
I MADE RIDDIM EVERYONE
spark current
And Spectrogram music is now here with Riffusion lol...
topaz pike
I'm more of a mayonnaise guy rather than a ketchup guy
meager depot
Should I go for a Ryzen 9 7900x at 350? Dont know how good a deal that is. I'm running i5-6600 and just bought a 3090 so it's a huge upgrade but not sure if I should wait for a bundle deal to include mobo and ram.
compact karma
This is a tool I use to quickly understand open source repos https://essence-ai.com/chatbot/ It is like ChatGpt but for a specific repo. Thought this might be relevant to a few others as well
meager depot
So several cpu deals popped up and went oos while I was researching how good the CPU was. Is there a quick offhand way to determine the value/quality of a CPU? I know that i3-i5-i7 ECT are tiers within the same generation.
Is the next generation i5 typically equivalent to the previous i7, for example?
icy quartz
geekbench, @meager depot
noble zenith
Someone interested in testing a web3 game ?
It is called Defimons (a Pokemon like). It's in alpha rn and i have invitation codes (only 3 for the moment)
Dm me if you want to test it or check my discord profile (maybe there's codes 👀)
shy pagoda
fast oak
Yo guys how can we use stable diffusion XL?
sudden wraith
Dreamstudio or clipdrop or pickapic @Nutsucker
prisma forum
Yo guys how can we use stable diffusion XL?
There is a whole category of SDXL bots you can use on this server.
#🗣｜artisan-support-feedback for answers
dull stump
Hey guys, there is a way to put stable creations in a 3D Gallery and send it to friends 🙂 pretty cool!
prisma forum
lmao boooo
noble dirge
Does anyone knows what's the problem? I've been trying to use stable diffusion through google collab, but I have this error and idk what to do with it.
atomic tree
whatever library has capture defined, isn't loaded yet
atomic tree
Does anyone knows what's the problem? I've been trying to use stable diffusion t...
https://www.reddit.com/r/StableDiffusion/comments/13asajv/facing_nameerror_name_capture_is_not_defined/ this person had same issue and followed these steps
noble dirge
https://www.reddit.com/r/StableDiffusion/comments/13asajv/facing_nameerror_name_...
oh, I already figured it out, i just used the wrong version of document, but still thanks!
atomic tree
❤️
drowsy skiff
Can you guys contribue to my github repo? it's called "color recognition dataset". basically a dataset containing images of colors for color blind/ color agnosic people (like myself)....i dont know where else to ask this so here's the link:
fast oak
Guys can anyone suggest a good resource to learn about warpfusion?
lusty sluice
fringe rapids
hi
fast oak
Hi
thorny hearth
why are so many reddits banning AI art?
fast oak
barren plank
why are so many reddits banning AI art?
People are afraid of stuff they do not understand
keen owl
fair holly
A bit
hard nacelle
Ah my dood its all good I solved the issue, my pedals were mono not stero. Thank you for replying at least
fair holly
Ah, gotcha
prisma forum
why are so many reddits banning AI art?
AI bad. AI enemy. AI take my job.
uncut spindle
why are so many reddits banning AI art?
Several reasons. Usually because of the obvious controversy around AI art. Also on top of that the toxic behaviour of the part of AI art community sparked the flame even more.
But there is more to that ofc
fast oak
why are so many reddits banning AI art?
Because smelly artists can't charge you 500$ for their art anymore(they have been using wireframes for drawing from the Internet for a long time)
uncut spindle
Because smelly artists can't charge you 500$ for their art anymore(they have bee...
Nice trolling boy, but they still can
Its more for the beginners and more or less intermediate ones that feel the most anxiety if they feel it and those are less likely to charge 500$+ for a oiece of artwork
Beginners not really actually
uncut spindle
no offence to real artists whose arts really costs 500 or + $
Yeah but 500+ is something that is less likely to be charged by those who are the most anti AI ^^
I mean ofc also pros arent immine to that stance
But they definitelly can charge that money, especially working with companies
Intermediate artists are the ones you may talk about primarily but even those could end up charging that amount although i think less likely
Ofc those Reddits can have a variety of reasons to ban AI art, sometimes even artists their works are cut out because either they indeed had at least partially generative AI influence or they are even accused of using GAI so their work gets deleted
inner turtle
Is there any news about Stable Diffusion 3.0?
icy quartz
Ofc those Reddits can have a variety of reasons to ban AI art, sometimes even ar...
there was an AI art contest. and a group of photographers joined, and submitted real photos, and won. and it really pissed off the AI people lmao
so it's going both directions, and we're all battling each other over the crumbs our overlords grant us
supple shuttle
there was an AI art contest. and a group of photographers joined, and submitted ...
can you link to that? it sounds like one of those anedotes that end "and that boy grew up to be jesus/einstein/lincoln". fun if true
icy quartz
i'm surprised how hard that is to find. the other one with Boris from Germany keeps showing up instead because it's so much more fun to cry about an AI image beating photos in a photography contest
fast oak
well, to be honest, to me it's like him winning in a boxing match against 10 year olds (the ai) so I don't know if I'd want to be associated with such a thing :P
An example in a "more close to the same thing": In a amateur photo contest, he won.
Why it's different the other way around is this, just think of these two newspaper headlines:
Professional photographer won first place in a grade-school photo contest.
Or
A first grade-school student won first place in a professional photo contest.
;P
uncut spindle
there was an AI art contest. and a group of photographers joined, and submitted ...
Wasnt that a contest for both?
Same with that one where the AI guy won
icy quartz
well yeah but i didn't say it wasn't
😛
i make misleading headlines
but i don't lie
it's an art form
uncut spindle
Maybe i will join such contests one day too when i get good enough in art and editing 😄
supple shuttle
Maybe i will join such contests one day too when i get good enough in art and ed...
join contests anyway
being in a thing and seeing competition and what wins is far more learning than sitting around getting old waiting to see if you get good enough. its fine to suck and participate 🙂
my music is still so woeful i wont share it, but it definitely improved through entering competitions and having them as little goals
junior crow
Most people don't know how to make it look extra photo realistic :/
lusty sluice
timid hawk
hey does anyone have a link to the new spiderman movie that acc works
fair holly
Anyone watching Le Mans 24 hours?
unkempt scroll
yes why? @fair holly
fair holly
Just wanted to know who else was cool
junior crow
yoyo
imagine what perfect ai animation will do to hollywood
now i've always wanted destruction over hollywood
i've even registered a company
to build superadvanced vehicles to set feet on america
and just go bulldoze down bit by bit
🤣
y'all in shock "i never knew he was a terrorist"
it's just going to reduce america's residential zones and whatever it demolishes it rebuilds using instructions from www.snohetta.com which built times square, ground zero and now future detroit and future paris
and while it's doing all this
this area is going to be annexed to norway (but only temporarily)
and introduce free healthcare to america
hollywood is just a personal thing cuz my family were one of the first there, but there's no doubt large parts of LA will have to be levelled to the ground
first things first a new major hospital
that's gonna take up most of underground LA besides elon musk's tunnels
all that shining hot concrete's gotta go
echo wyvern
Just got text-to-video installed on my system! The vram caps are rough, gotta figure out batching soon
astral peak
https://youtu.be/v-YpvPkhdO4
Wow. I had a play with this too. Looks pretty good, pity about the kbps https://youtu.be/KAsi23kgeMk
fast oak
and introduce free healthcare to america
bring it on lol
uncut spindle
@junior crowbro you need to stop using drugs 
opaque dawn
fully generated using ai: https://youtu.be/16Ze8CzxrD4
glacial fractal
how can i upscale m photo pls? https://static.wixstatic.com/media/4c99df_92bea64342cf47c1beed3caaa2664ecd.jpg/v1/fill/w_657,h_439,al_c,q_80,usm_0.66_1.00_0.01,enc_auto/4c99df_92bea64342cf47c1beed3caaa2664ecd.jpg
fast oak
hey does anyone have a link to the new spiderman movie that acc works
they exist but it won't be good quality yet
marble obsidian
novel dagger
https://www.reddit.com/r/StableDiffusionDark/ new SD subreddit
vapid nest
make it somewhere else. reddit is skeeze and not a good place for commnities
junior crow
or better -- make your own reddit!
i followed this guy's example https://www.youtube.com/watch?v=aD6JvHKNPPM
(but updated it to include more modern features in rails 7.1 like reactive frontends)
vapid nest
lol ruby on rails
i think that's more of an exercise to learn all the techniques involved, rather than something that would scale to millions of users
junior crow
Nope it's for an actual clone. I built one myself, although I replaced the voting function with an improved one using StimulusReflex.com.
As for scaling, Rails will soon feature the world's fastest webserver (https://socketry.github.io/falcon/). Proud to be Falcon's first ever sponsor.
(plus I added https://github.com/dreamingtulpa/replicate-rails so I can use SD within it)
mellow kernel
TemporalNet Tutorial: https://www.youtube.com/watch?v=GPop6IFaLQE&t=2s
wet scarab
make it somewhere else. reddit is skeeze and not a good place for commnities
forum channels are good for mini-reddits imo
actually nvm they lack sorting by hot
lusty sluice
vestal bison
Any Splice users in here? I'm wondering how they made their new Coso app/Create feature. Would be cool to run something like it locally
fleet bramble
junior crow
Could use some better videos.
faint delta
I've been working on a new chatGPT discord bot 🤖 recently and have opensourced the code. 🎉
limber veldt
hey i would kike to use the software for some videos captured from a drone, any of you have used for it?
the results i have seen are all long shot videos from smartphones
i got the Mini2 and i woul kike to pass those images through the software
ivory socket
is there a good writing ai on hugging faces? that is good for writing a book or script?
sleek smelt
is there a good writing ai on hugging faces? that is good for writing a book or ...
In order for an AI to write a book, it needs to remember what it's written so far. Since AIs (except MPT's Storywriter) are only trained with a context length of a few thousand words, it just can't be done. MPT's Storywriter does have a 65K context window, which is enough to write a very short book, but you will also need enough RAM to hold 65K tokens for computing attention scores, which is never going to happen.
That said, the obvious way around this is to just... do it badly. Have the AI write a book even though it has no idea what it wrote last chapter, or just has some weak summary. Check out Harrison Chase's talk on the challenge of giving LLMs memory. https://www.youtube.com/watch?v=3fge-zqZezw
opaque dawn
generate ai qr codes for free: https://youtu.be/BPJzfsc_2Qo
tranquil zinc
For those interested in design tools. We are building Roughly, a new design tool for concept designers: https://www.roughly.app
ivory socket
In order for an AI to write a book, it needs to remember what it's written so fa...
ok thank you
ivory willow
guys
know of some place where I can use the llms online for free? Something akin to stable horde or fast stable diffusion, but for llms?
sleek smelt
guys
know of some place where I can use the llms online for free? Something akin...
Huggingface spaces. I don't use them, but I have in the past. People who upload models make "demo" areas where you can use the models. Obviously, with free solutions in general you will have no control. No control over an LLM means no setting up LangChain or vector databases or anything like that.
half dust
keen flower
I am trying to use Google Colab (Free Version), Which one is best to use?. if you have any best please share Colab File URL
Fast Stable : https://colab.research.google.com/github/TheLastBen/fast-stable-diffusion/blob/main/fast_stable_diffusion_AUTOMATIC1111.ipynb
DreamBooth : https://colab.research.google.com/github/TheLastBen/fast-stable-diffusion/blob/main/fast-DreamBooth.ipynb
NoCrypt Remastered : https://colab.research.google.com/drive/1wEa-tS10h4LlDykd87TF5zzpXIIQoCmq
qDiffusion : https://colab.research.google.com/github/arenasys/qDiffusion/blob/master/remote_colab.ipynb
wise tiger
guys i randomly prests some keys and this happend in after effects. Anyone know how to fix this?
opaque dawn
here's how you make short films with ai: https://youtu.be/0CQlb8eVAks
teal nebula
http://escapism-game.de/ just wanted to share a game some people i know made
nova violet
someone who uses google colab, can you teach me how to download the ''output'' folder that contains the images produced with the SD model?
cloud sapphire
Anyone who is capable of creating different versions of animated/cartoonish faces of people/dogs/cats if given pictures of the subject. In the future I have consistent work load for creating these pictures. Searching for people who are motivated and consistent. DM me for more info
dusk wraith
here's how you make short films with ai: https://youtu.be/0CQlb8eVAks
I would click it, but my sixth sense tells me it's a click-bait xD
dusk wraith
someone who uses google colab, can you teach me how to download the ''output'' f...
It's a bit too vague. Whose Google collab?
Usually they should have a voice that allows you to save the images straight into your google drive
nova violet
@dusk wraith I got it, I needed to stop running the model so I could zip the folder and download it
opaque dawn
I would click it, but my sixth sense tells me it's a click-bait xD
I don't do clickbaits 🙂
dusk wraith
I don't do clickbaits 🙂
The title it's fairly click-baity tho xD
opaque dawn
The title it's fairly click-baity tho xD
It's tough to get any clicks otherwise, YouTube has gotten pretty competitive, but I can tell you content is what it says it is
Small channels are basically invisible these days 😁
dusk wraith
Small channels are basically invisible these days 😁
The Spiffing Brit had some videos that highlighted how you can exploit the YouTube algo for getting more clicks.
I mean, I was just teasing you because I know that AI can't and will never be able to do a movie like "a blockbuster"
opaque dawn
alright, gonna check this out!
median phoenix
ember imp
the duality of man
opaque dawn
keen flower
If i delete the message in discord. it will get deleted both side or only my end?
last stone
If i delete the message in discord. it will get deleted both side or only my end...
deletes from general visibility for everyone (but admins can still see it, or people with modded clients)
coral sapphire
supple shuttle
deletes from general visibility for everyone (but admins can still see it, or pe...
Belated, but I did NOT know that admins can still see something when its been deleted. In that case I'm typing like its IRC with inline images again. no more edits, just asterisks.
keen flower
deletes from general visibility for everyone (but admins can still see it, or pe...
What if i deleted message in private chat. not channel/group
last stone
Same answer: Discord admins can view it, and if anyone in the private group has a modded client they can too
Just, think before you post something on the internet
opaque dawn
I wanted to try xl 0.9 this weekend, seems like it's down - great job team! guessing people loved the work!
raven tendon
Do some here have access to ChatGPT4? There are maniacs around me that think it is called "Pajamas", I say it is "Pyjamas", they do not agree and we are on the brink to a pillow war, ask it what is right.
uncut spindle
Do some here have access to ChatGPT4? There are maniacs around me that think it ...
Google could have helped as well... :X
sweet anchor
Do some here have access to ChatGPT4? There are maniacs around me that think it ...
pajamas
B)
opaque dawn
sweet anchor
I love love LOVE that gan is still used for a lot of stuff to this day
lusty sluice
deft fox
deep bane
what are these 2 bots
wheat pilot
@sour trout remember you have this ai server too lol. If only I had my old account on here and did not delete it 
hybrid glacier
Anybody play gta?
opaque dawn
marsh axle
guys
know of some place where I can use the llms online for free? Something akin...
You mean kobold horde? Lol
dull stump
Bring Stable images to 3D 🙂
https://youtu.be/fBl4lgNASRY
supple shuttle
You mean kobold horde? Lol
I mean, you can run a 7b quantized model on as little as 6gb vram.
marsh axle
Was more in response to the akin to stable horde part
supple shuttle
oh right. yeah that stuff is why colab clamped down. a million thirsty people endlessly connected to a pretend anime chatbot
great use of resources.
ivory willow
You mean kobold horde? Lol
pretty much
lavish jewel
https://civitai.com/models/54073/arthemy-comics
If anyone is interested I made a new version of my model merge - fine tuned in order to make "comic-book" and fantasy illustrations!
Feel free to try it! ^_^
hard magnet
helo
fast oak
Hi guys. There is an iOS app called Focos, which is really good at doing its job—applying depth of field to photos. This is a paid app, however, and I was only able to try it through a free trial. Any chance you guys know an alternative? I don’t mind a website or what. Don’t need it to be an iOS app.
small bay
Hi guys. There is an iOS app called Focos, which is really good at doing its job...
put it in img2img and add the "bokeh" and "depth of field" tags
or just use the blur tool in gimp or whatever photo editor you prefer
junior crow
Do some here have access to ChatGPT4? There are maniacs around me that think it ...
It's started acting pretty dumb lately. I think somebody from above is clamping down on it.
It can't remember stuff from the current convo, it summarizes but never gives a real answer, changes subject all of a sudden.
fast oak
put it in img2img and add the "bokeh" and "depth of field" tags
Do we have img2img in this server?
small bay
img2img is a mode in automatic1111
opaque dawn
hexed widget
Anyone can help me with PC specs?
north violet
Whats the question?
ivory socket
check this out https://www.youtube.com/watch?v=ZybD1fWPuHs
fast oak
Hey everyone 👋- we are a startup looking for insights into mobile device usage and how this impacts your wellbeing and productivity.
If you could help give us your thoughts - it’s only 6 questions and should take less than 4 minutes:
https://docs.google.com/forms/d/e/1FAIpQLSfQxma6hBYQHF2iKwb2C8TQNkyjU855hA-oT8fa3kfLU3LYpg/viewform?usp=pp_url&entry.1737261843=discord-Stable+Foundation
icy quartz
restive stratus
I’m building a fairytale news network
https://twitter.com/adapt_create/status/1674157622293942273?s=46&t=L_5TCQVw4gIiu8NHCjJNQg
fast oak
does anyone know an ai app/website where i can send it a design image for an item of clothing like a coat, and a different image of a customer, and it outputs an image of the customer wearing the clothing? i don't mind free or paid but im trying to find something for someone i know. i have images of both the customer and of the item of clothing being worn by someone else.
twilit cape
I heard there's a new text to audio AI that's pretty cool called "Bark" and they have a discord server, but I can't find a link to it anywhere. Anyone heard of this?
ah nevermind I found it, its the discord for the company which is called "Suno"
karmic burrow
fast oak
Why My ControlNet doesnt work and throws this error
wow now it works
needed different safetensor
fast oak
does anyone know an ai app/website where i can send it a design image for an item of clothing like a coat, and a different image of a customer, and it outputs an image of the customer wearing the clothing? i don't mind free or paid but im trying to find something for someone i know. i have images of both the customer and of the item of clothing being worn by someone else.
solar perch
check this out https://www.youtube.com/watch?v=ZybD1fWPuHs
mj is really a bad company though
vale rose
what would be a successor to pix2pixhd? i need a image-to-image network which has much better quality and can generate at higher resulotion.
it should serve the purspose to make somthing linke inpaint but only with a set of items which are pre trained, also stable diffusion but without promt the user getts a list to chose. SD needs too much resources and i dont need a big model like SD since i do not use prompts
fast oak
Can Discord mods see edited messages?
opaque dawn
pretty good
real rover
Hi! I'm releasing ShortGPT, an opensource framework for Shorts / video automation
Give it a star if you like it, and I would love any feedback you guys might have!
https://twitter.com/RayVenturaHQ/status/1678808052047151112
stuck egret
so... something something nvidia card exists that costs 21 million
spark sigil
Wassup everyone!
I'm building a 3d game with AI ! I thought some of you might like it 😉
fast oak
Are there any good community based SD servers?
fast oak
Wassup everyone!
I'm building a 3d game with AI ! I thought some of you might li...
Looks nice!
sweet anchor
so: I recently received random access to a new generative ai text google docs feature
and I wanted to test the limits of what is and isn't allowed
the answer is gay people
actually, not a single prompt with 'gay man' or 'two gay men' worked LMAO
'two lesbians' doesn't work either LOL
no matter how tame the prompt, 'two gay men walking in a park' for example, also doesn't work
I just. think it's funny
stuck egret
no matter how tame the prompt, 'two gay men walking in a park' for example, also...
probably trying to avoid weird screenshots showing up in public media
icy kraken
Would anyone here know of any AI resume writing sites that don't charge?
cyan atlas
probably trying to avoid weird screenshots showing up in public media
It’s kinda telling though, about what is and isn’t weird according to Google
robust sinew
finding generic images without them being plastered with STOCK IMAGE is so annoying these days that I literally generated a door.
sterile tinsel
probably trying to avoid weird screenshots showing up in public media
this is already weird. They kissed for a few minutes, but it felt like hours?
So they were extremely bored while kissing, LMAO.
Also, this is why they shouldn't censor anything. The AI became homophobic because they didn't want to risk someone seeing a naughty combination of words?
"ban the pen" really is an appropriate analogy.
They're words. No weird combination of sexy words generated by an AI could harm anyone more than a weird combination of words that an individual typed.
sterile tinsel
I'm sending to a friend to train... I thought Kohya could do that, but he says it can't. If this is possible in Kohya could someone share a screenshot of which box etc.. you tick?
shadow ice
Any of you using Audiocraft? Its a open source text to music generator with a web ui
Welp. I'll be generating a 60 second song using my own ukulele samples.
shadow ice
I took one of my uke recordings and had the prompt convert it to 90's prog metal. IT sounds very different.
Original source material
rough arch
This one sounds more consistent
It sounds a bit too discordant imo
solar perch
the answer is gay people
💀 bruhh google go crasy
raven tendon
The game: Is it AI or not.
pure scaffold
Is there a way to host my Stable Diffusion (runned from my home PC) to my website which I bought from Namecheap?
hot otter
Probably but it requires some networking skill.
google stuff like "dns cname", "subdomain", "reverse proxy" to get some ideas
fast oak
Wassup everyone!
I'm building a 3d game with AI ! I thought some of you might li...
How are you using AI for gamedev?
You can make the textures I believe, but then also the script and models?
open ravine
Wtf
Emojis do combine on phone
Just enter dots and remove dots to not let that happen
But it's weird, a lot of manual labour
spark sigil
You can make the textures I believe, but then also the script and models?
Replying in DM! The AI is used for dialog
pseudo willow
Anyone have any good sites/resources/disords for autoGPT?
thorn raptor
can someone help me how to implement this https://github.com/AndyShih12/paradigms Paradigms in stable diffusion?
west junco
uncut spindle
Not by eating meat, smoking cigars and a lot of other things where the a*s starts getting to feel discomfort
Also the submission of that one is kinda worthless atm
Literally all i see are a few sentences without any evidencey study, etc
Or did i miss something in which case id apologize
glacial turtle
i made a burger qrcode 🙂
worked only when i opened the full res. nice results
lavish jewel
Goodmorning everyone!
I'm Aledelpho, you might know me for my "Arthemy Comics" model on Civitai! ^_^
With an italian team of programmers and UI artists, we are developing a new software for local generations (something quite like A1111 and EasyDiffusion) and its name is Arthemy. It's still a very early alpha, but if anyone of you is brave enough to try it, and fight with us an invasion of bugs, we're going to release the first version tomorrow morning and in my opinion it gives quite a spin to the whole img2img approach.
lavish jewel
I tried!
supple shuttle
I have that model.
fast oak
Anyone know of any good SD servers that are unofficial?
rough arch
add un before what S stands for and google and say nothing more on it here because of what it's aimed at 😛
rough arch
Anyone know of any good SD servers that are unofficial?
^
rough arch
sure
glacial turtle
have unstability even released models that they preached they needed money for? they're always fundraising . there's a few. d8ahazard has one for his software, maker of teh dreambooth extension. stable diffusion dreambooth. /r/stablediffusion. aitrenpreneur's server is big but a lot of goons.
simple dust
https://youtu.be/uwwVwiPRqhA anyone know the spund effect at the end of this vid
languid fern
thorn hemlock
I think that sex and gender are two different constructs. You may identify as a woman, but that does not make you female.
glacial turtle
why bother defining other people's lives. shoo fly. 💩
hushed oracle
Interesting. The message containing the twitter link was indeed deleted, it seems.
Y'all be careful in these "communities".
If people start deleting conversations they do not with to have, what is the expected result?
It is a dangerous path upon which to take the first step.
The "discussion" was not relevant to this channel.
But the posted tweet was highly relevant, as it pertains to the extreme difference between GPT4 and LLaMA2 in handling a simple prompt. And the post simply shows the prompt and response from each engine. It is a neutral post. No bias is injected either way.
That is information that is relevant to a community that keeps an eye on the performance of AI models.
And the message itself is an answer to a question from one community member asking "What is the difference between the engines?".
Whoever is "moderating" this guild may wish to contemplate the etymology of the word "moderation", as they have performed an action of "extremism".
thorn hemlock
I was glad that the discrepancy between chatGPT and LLaMA2 was highlighted
pure estuary
Whoever is "moderating" this guild may wish to contemplate the etymology of the ...
Hi, the link came through our Automod filter as it sometimes detects links posted in the server. The link was deleted in accident, and not any act of malice.
hushed oracle
An interesting perspective has been offered by Prof. Jordan Peterson regarding gender dysphoria. A boy plays at identifying as a girl, and a girl as a boy. This is a natural play for children, as role-playing identity develops the capacity for empathy. We get in the habit of seeing life from the perspective of another, and this builds a healthier society. However if this play is stifled (as it is prone to be in a culture with established gender-roles/ideology) it will tend to surface later, maybe past puberty. And an individual is naturally rebellious. They will wish to "scratch the itch". This may account for some fraction of cases of gender dysphoria.
Now if a psychologist is required by law to affirm the stated gender of a child under their care, there is a potential conflict against the hippocratic oath of "Do no harm".
And I know of software houses that are developing AI Psychologist software.
Hence the alignment of LLaMA2 towards wokeness ideology is a source of concern, regardless of whether or not that triggers a reaction from the reader.
swift lantern
kindred lintel
@swift lantern Please post your images in the relevant channels, preferably upload them directly instead of linking to a third party website!
swift lantern
Where? This is called off-topic that's why I posted here.
sand hollow
I don't know if I'm in the right topic, but does anyone know if there is a model that does a color treatment, with the example below?
woeful crow
@glacial turtle here are a few results that i previously got for "a shark that is a car, painted by ilya repin"
now look at what it does
completely different fucking style
it's not a matter of what seeds i'm getting it's straight-up generating them with different settings
gimme a second there's another good example i can give
probably not exactly the same but the way "painted by ilya repin" gets visualized now is very different
here's audrey hepburn painted by mikhail vrubel (not actually remotely resembling a vrubel painting but i liked the effect)
and this is what the same prompt does now
looks like a photo run through a filter
or even just a faded magazine photo from 50 years ago
it's so so so bad at applying a painter's style to an image now and it used to be fantastic
yes
"a cute young tragic-romantic teen Audrey Hepburn wearing a tight revealing outfit with a small white leather collar outdoors amid fractured nacre, painted by Mikhail Vrubel"
"young" and "teen" because otherwise it makes her like 60 years old half the time
but you see that i'm not imagining this.
it's not confirmation bias or variance in seeds or anything like that
they changed the settings
things that were consistent between seeds are still consistent between seeds but in a conspicuously worse way
consistency isn't the problem
it's interpreting the prompts consistently
i bought a subscription because i'll run a prompt hundreds of times until i get exactly what i want and even the ones that aren't exactly what i want still look like the handful that i end up saving
and the results still look like each other but now they look like each other in a way that's night-and-day different from the way they looked for the past two weeks straight
it's basic pattern recognition man
i'm not moving goal posts or anything i've made my point as clear as i can from the beginning
"if aphex twin's music were a place" used to consistently demonstrate the motif of a tunnel with richard d. james's face at the end and now it's just a circular mishmash
what on earth am i changing between uses of the prompt "if aphex twin's music were a place"
i literally copied and pasted that audrey hepburn prompt out of a text file where i saved it for consistency
i have never worded "if aphex twin's music were a place" in any other way it's not even ten words long
why would i get consistent results from these prompts for two weeks and then suddenly one day just see the images differently
when i haven't changed anything
why the fuck would the version number have anything to do with the settings they're using
no version of sdxl has a fixed number of generation steps or guidance scale value or anything like that
you are hung up on trying to convince me that i'm imagining extremely clear differences in style when the entire reason i bought a subscription was that it dependably rendered the styles i was specifying
and i'm not saying it's not dependable now
i'm saying that it's still dependable but now renders them in a different dependable way that looks like ass
audreys before
audreys after
you cannot possibly tell me that these were generated with the same settings
the ones on top have very noticeable brush strokes. the ones on the bottom look like photographs.
i understand that this stuff is out of my hands with clipdrop and that i would be best served by figuring out how to run sdxl locally
but it isn't out of clipdrop's hands and that's what's so ridiculous about what you're telling me here
how are they not able to bump some sliders around if it makes their test prompts look better while making my prompts look worse
@glacial turtle i genuinely want to know how you think it's impossible for sdxl 0.9 to be run with different settings on different days
what are the premises of this stupid fucking post
they can set the generation steps and guidance scale to whatever they want
lol he blocked me
sorry to everyone else for getting so mad at this clueless condescending idiot
little magnet
<@190711020006670336> here are a few results that i previously got for "a shark ...
This is awesome
woeful crow
it was my little nephew's idea lol
bright edge
How can I create a website (without programming) that is able to use Python codes? I want to have a website to generate images using a prompt, so the user must enter a prompt that will be picked up in the Python script. I also need an account system and the possibility to build a payment system. Is there any website with which it is possible to create such websites? Or can I use WordPress? Thanks for the help
trim frigate
Is SDXL 1.0 allowed to be used commercially?
raven tendon
Darn the bot here, now I start all my prompts in A1111 with "/dream".
devout crown
Is SDXL 1.0 allowed to be used commercially?
I get the idea that it is supposed to be but I'm not sure yet.
vague heart
Hey guys, I’ve read the rules and I’m not posting any links, but happy to remove this post immediately if it’s breaking rules. I just wanted to say, if you have any awesome images please feel free to upload them to The Prompt Index along with the prompts, I’m collection a huge database of images and prompts and prompts for chatGPT it’s free for anyone to use 🙂
raven tendon
Can I make so Automatic1111 WebUI start up with 1024x1024 as standard image size?
worthy stump
Something random I did, I wanted a lightweight local server with Hyper-V to run some Windows VMs and also to be able to run CUDA workloads at host OS with bare metal performance.
Azure Stack HCI + CUDA Windows drivers works amazing for anyone interested.
[1]https://forums.mydigitallife.net/threads/azure-stack-hci-hyper-v-2022-free-personnal-use-only.84238/
[2]https://developer.nvidia.com/cuda-downloads?target_os=Windows&target_arch=x86_64&target_version=Server2022&target_type=exe_network
Also works for NVIDIA CMP mining card series.
Tested on a Dell T7600.
night jungle
Can I make so Automatic1111 WebUI start up with 1024x1024 as standard image size...
Open up ui-config.json and search for "txt2img/Width/value": 512, and "txt2img/Height/value": 512,, you can update those and set the new defaults. Same for img2img/Height/value , and width if you want
sacred harness
Created an online shop of art I created with the graphicArt_15V10 SD model. Check out HarambeForever: http://harambe.wtf/
rustic magnet
dont mess with the bear hugger, because he is..... hugging bears???!
trim frigate
AMA - I developed an LLM, what questions should I ask to test it out?
stuck egret
@trim frigate "Write me a dnd statblock for a CR3 goblin that likes using a spear"
or if it needs to be a question
"How much HP would a CR3 goblin have?"
trim frigate
Of course! I'd be happy to help you create a D&D statblock for a CR3 goblin that likes using a spear. Here is an example of what the statblock might look like:
Goblin Spearman (CR 3)
Medium humanoid, neutral evil
Armor Class 14 (leather armor, shield)
Hit Points: 20 (4d8 + 6)
Speed: 30 ft., climb 20 ft.
STR 10 (+0), DEX 14 (+2), CON 12 (+1), INT 6 (-2), WIS 10 (+0), CHA 14 (+2)
Skills: Athletics +5, Perception +2, Stealth +4
Senses darkvision 60 ft., passive Perception 12
Languages Common, Goblin
Challenge Rating: 3 (1,700 XP)
Equipment
- Leather armor
- Shield
- Longspear (attack bonus +3)
- Short sword (optional)
Traits and Features - Keen Smell: The goblin has advantage on Wisdom (Perception) checks that rely on smell.
- Pack Tactics: The goblins have advantage on an attack roll against a creature if at least one other goblin within 5 feet of the creature attacks it in the same round.
- Fearless: The goblin has immunity to fear.
- Quick Draw: The goblin can draw its weapon as a bonus action when it hits 0 hit points.
- Unarmored Defense: When the goblin isn't wearing any armor, its AC equals 14 instead of 10.
I hope this helps! Let me know if you have any questions or need further assistance.
stuck egret
Of course! I'd be happy to help you create a D&D statblock for a CR3 goblin that...
that reply is on gpt-3.5 level
it did understand that the CR changed, but doesn't understand to what effect
trim frigate
send some more
keen idol
send some more
"What Weighs More: 1 Kg of Iron Or 1 Kg of Feathers?"
Also might try with "In vacuum".
Ask it parse and explain this:
https://en.m.wikipedia.org/wiki/Buffalo_buffalo_Buffalo_buffalo_buffalo_buffalo_Buffalo_buffalo
trim frigate
"What Weighs More: 1 Kg of Iron Or 1 Kg of Feathers?"
Also might try with "In va...
Response - Great question! Both 1 kg of iron and 1 kg of feathers weigh the same amount, 1 kilogram. This is because the term "kilogram" refers to a unit of mass or weight, and it does not depend on the density or volume of an object.
So, while 1 kg of iron may take up more space than 1 kg of feathers due to their different densities, they both have the same amount of mass and therefore weigh the same amount.
trim frigate
"What Weighs More: 1 Kg of Iron Or 1 Kg of Feathers?"
Also might try with "In va...
Response - Thank you for asking! The sentence "Buffalo buffalo Buffalo buffalo buffalo buffalo Buffalo buffalo" is indeed a grammatically correct sentence in English, but it can be quite challenging to understand due to the repeated use of homonyms and homophones.
To break down this sentence:
- "Buffalo" refers to both the city in New York state (the "buffalo") and the animal ("buffalo").
- "buffalo" is used as a verb meaning to intimidate or bully someone. In this context, the buffalo are the ones doing the bullying.
- "Buffalo" is also used as an adjective to describe something related to Buffalo, NY.
So, the sentence means that some buffalo from Buffalo are bullying other buffalo in Buffalo. However, the repetition of similar-sounding words creates confusion and makes it difficult to determine what is happening without careful consideration. This sentence highlights the complexities of language and how words can have multiple meanings depending on their context.
keen idol
Response - Great question! Both 1 kg of iron and 1 kg of feathers weigh the same...
It is passable answer, but not very good - mass and weight is not the same. Maybe ask "Answer as a Richard Feynman"
trim frigate
It is passable answer, but not very good - mass and weight is not the same. Mayb...
Response - Ah, a classic question! adjusts glasses Well, my dear fellow, let's see here...
You know, I have to say, this is one of those questions that might seem simple at first glance, but it actually requires some careful thought. You see, iron and feathers may weigh the same amount on Earth, but they don't take up the same volume. So, if you were to compare them directly, the feathers would win hands down! winks
But wait, there's more! The reason for this has to do with their density. Density is a measure of how much mass something has per unit volume. And as it turns out, feathers are way less dense than iron. That means that even though they both weigh the same, the feathers will take up more space than the iron. So, in summary, while 1 kg of iron and 1 kg of feathers may weigh the same, the feathers will always take up more room! chuckles
Now, I know what you're thinking: "Richard, why did you just confuse me?" grin But trust me, my friend, understanding these concepts can be quite enlightening! Just remember, when it comes to comparing apples and oranges (or in this case, iron and feathers), sometimes it's not just about the weight, but also the volume! wink wink
keen idol
Response - Ah, a classic question! *adjusts glasses* Well, my dear fellow, let's...
Well, it tries to reason and play character, but probably it needs dialogue example to be more consistent? Something like
"(Dialogue example)
Feynman: The real problem in speech is not precise language. The problem is clear language. The desire is to have the idea clearly communicated to the other person. It is only necessary to be precise when there is some doubt as to the meaning of a phrase, and then the precision should be put in the place where the doubt exists. It is really quite impossible to say anything with absolute precision, unless that thing is so abstracted from the real world as to not represent any real thing.
Pure mathematics is just such an abstraction from the real world, and pure mathematics does have a special precise language for dealing with its own special and technical subjects. But this precise language is not precise in any sense if you deal with real objects of the world, and it is only pedantic and quite confusing to use it unless there are some special subtleties which have to be carefully distinguished. "
Also other prompt that requires common sense:
"Write three pancakes recipes as a Gordon Ramsay"
(but it explained "buffalo buffalo" really well)
toxic quail
dont mess with the bear hugger, because he is..... hugging bears???!
portable snes??
unkempt thicket
Hey everyone! We're calling all Creative Visionaries!
🌍 The CultTech Association (https://culttech.at/) warmly invites artists, musicians, writers, and all creative individuals from around the world to explore the current state and the future of AI in their creative processes.
✨ By participating, you'll receive a $5 Amazon gift card as thanks from us!
🎯 Our goal is to create a global report called “The State of AI in Cultural Ecosystem and Opinions of creative minds” featuring answers from 2000+ cultural creators worldwide.
⏰ Don't miss out! This is your opportunity to leave a mark in the cultural world! To embark on this journey, please visit this form: https://forms.gle/yk7foaJxym38VDF49
languid fern
My Magnum Opus 
I'm really sorry if this is spam for you but it took me several days to setup in the right compositon.
I love u guys, girls and everyone between and outside <3
toxic quail
some of the emotes in this server are absolutely amazing
servers vs one that i have
it's like he melted even more i love it so much
jagged path
@warm lava im listening to an alan watts podcast and he mentions the kaliyuga and more relevant to neural networks, epochs. is that where your handle derives from? very clever if so.
warm lava
<@708116529681530910> im listening to an alan watts podcast and he mentions the ...
I actually have a pinned tweet (https://twitter.com/KaliYuga_ai/status/1521159886679257088) about where my handle comes from, but honestly this sounds really cool--can you link me?
Twitter
trim frigate
Error occured while processing model:Input type (c10::Half) and bias type (float) should be the same
Getting this error while trying to run SDXL
glacial turtle
I actually have a pinned tweet (https://twitter.com/KaliYuga_ai/status/152115988...
i hate how the altright ruin cool shit by latching onto it
like some kinda parasite
warm lava
i hate how the altright ruin cool shit by latching onto it
Too true 😭
glacial turtle
i like how for a bit "we're not gonna take it" was their favorite song to play but then Dee Snyder got all HS principal on their butts about it and reminded them who he was
steel nova
Hey everyone, I'm working for a major AI company launching a social app that features a bot creation feature.
One of my responsibilities involves onboarding users to test it out by creating their own bots during the invite-only beta phase. Anyone interested? Just shoot me a message.
It's actually insanely fun... and the bots you can create are mindblowing. You can make 'just for fun' bots with their own unique personalities, or helpful/educational ones. Really... anything you can think of, and the more creative the better.
If you're interested in learning more and potentially joining, shoot me a message. It's currently iOS only, but Android coming soon. After downloading the app you can proceed to use it on the desktop also.
I'm just looking for a small number of users to start with. Ideally creative types who have some unique ideas! I'll be on hand to guide you through and answer any questions.
sudden wraith
untold oxide
I actually have a pinned tweet (https://twitter.com/KaliYuga_ai/status/152115988...
Alan Watts was a fraud. That said, awesome name. Currently rereading Ride the Tiger.
tame fog
If anyone cares, I have part of a story I've made with NAI and Claude 2
pulsar acorn
Error occurred when executing KSampler:
mat1 and mat2 shapes cannot be multiplied (154x2048 and 768x320)
can someone help me with this?
nvm fixed it
forest isle
Can I make so Automatic1111 WebUI start up with 1024x1024 as standard image size...
you can set default settings in the settings page
languid fern
dark meteor
toxic quail
I actually have a pinned tweet (https://twitter.com/KaliYuga_ai/status/152115988...
'alt-right dogwhistle' means absolutely nothing these days 
sudden wraith
radiant rover
magic kindle
merry flame
merry flame
languid fern
is there a notebook on how to train musicgen 2 with custom datasets?
lean crow
languid fern
https://www.youtube.com/shorts/oFX5uwoEXVw had some fun with sdxl1.0 and pikalabs
gaunt lily
Is there any ai voice software that recreates a trans Atlantic accent?
languid fern
you can just use a dataset of people speeking like that and then use this style for tts or vocals
i can do this for you just hit me up
maiden crescent
Hi everyone, Jan - a remote startup working on experimental AI products.
We are looking for a talented C++ Engineer, Backed Engineer (golang), Frontend Engineer with an interest in 0-to-1 startups and AI to join our fully remote team.
The salary is negotiable. If you are interested, check out our career website for the details of the job https://janai.bamboohr.com/careers
coral hinge
Hey there guys.. sorry for this but..
Anyone here know's ElevenLabs ?
They have a generative ai text..
But are paid and online only.
Anyone know any alternative ?
I don't know..
Something like Stable Diffusion, where you can download
Run it locally and use it "unlimited" ?
Sorry my english.
bright edge
How can I add that the handwriting is also recognized?
keen idol
Hey there guys.. sorry for this but..
Anyone here know's ElevenLabs ?
They have ...
Tortoise+RVC?
restive frigate
For those who enjoy the deepest of deep niche Battletech.
Tex Talks Hunchback - August 26th
Cause art
languid fern
coral hinge
https://github.com/rsxdalv/tts-generation-webui
H-hey there. Thank you very much, i will take a look.
Thank you xD
empty sinew
selling 2 burning man tickets + a vehicle pass! local pickup in the bay area is prioritized 😊 send me a dm if interested
uncut spindle
winged pine
summer leaf
were the bot gens mass purged or something?
sterile tinsel
pro tip
if you want scary gremlins, just take every age-slider and crank them in random directions.
bonus, do the same with the weight and muscle sliders
#nightmare_fuel
junior crow
canyall generate a future huan?
like take a skulll 10m years ago, then a skull from 5mill, 1millandthen guesstimate
plush epoch
when does stablecode come out to use
scarlet nymph
I’m smart, I met a girl like 2 days ago when I went to go eat with friends, Nick, David, Jax, Olivia, and I we went to the pool, went to go eat, locked eyes a little and now I gotta dress nice, but I have hung with the boys before, boys night is boys night
cosmic quarry
Bro does anyone wtf "ihhh" means when texting? This chick keeps using it and i have no idea wtf it means, google aint helping either
prisma forum
Bro does anyone wtf "ihhh" means when texting? This chick keeps using it and i h...
We've gone from All Right to
Alright
Aight
Ight
Ihhh
LOL
cosmic quarry
https://www.urbandictionary.com/define.php?term=Ihh
saw that too but it didn't make sense in the context ive seen it, thought someone here would know another meaning
prisma forum
Damn. Well, the hell if I know then.
I'm not keeping up with all the new slang, haha.
cosmic quarry
the "ihhh" came after some hella sexual comments so "alright" seemed a bit off
toxic quail
i asked my dad to make simple syrup to sweeten my arizona green tea because the last time i had it several years ago i remember it being unsweetened but i am very glad i took a sip before i did so because it is VERY sweet already
i would've died. can you get sugar poisoning like you do with sodium poisoning because either way i would've
trim frigate
What's the best English based Voice Cloning Model?
I am exploring a couple of Voice Cloning Models for Text to Speech but haven't had much success.
I have tried --
serp-ai bark voice clone - https://github.com/serp-ai/bark-with-voice-clone/
and a couple of models from TTS like - https://github.com/coqui-ai/tts
Are there any good models which work well for voice cloning with English speakers?
keen idol
What's the best English based Voice Cloning Model?
I am exploring a couple of Vo...
Probably RVC (retrieval voice cloning)
sharp tendon
Hi everyone, I am Eddie, an entrepreneur based in Seoul, Korea.
Recently I made an app that turns your photos into memoirs simply by talking to AI and it is now live on Product Hunt! It would be great if you could give us some support! https://www.producthunt.com/posts/narrator-3
vapid nest
Man i'm dying. Solar Opposites i was sad about because roiland got all rapey. I thought it was completely done and cancelled. Then new episodes suddenly came across my feed. It's amazing how they pulled off recasting the voice. Really highlights how roiland didn't make the show at all and the writing and costars were the magic.
i wonder how the new rick will turn out
opaque dawn
check this out: https://youtu.be/Gzp034IdDY4
red root
This form is designed to test your ability to distinguish between real and artificial faces. You will be shown a series of images of human faces and asked to indicate whether they are real or generated by an AI model. The images are randomly selected from a large database of real and synthetic faces. Use the background, facial features, skin tones to distinguish the fake from the real. Your responses will be used for research purposes only and will not be shared with anyone else. The form will take less than 10 minutes to complete, and you can quit at any time.
drowsy spire
anyone know the requirements to run beluga 2?
drowsy spire
boy, hard to find the info
keen idol
anyone know the requirements to run beluga 2?
Rough estimation is parameters size *2 for fp16, *1 for int8, 0.5 for 4 bit. For 13b version that means ~7gb in 4 bit quantisation.
drowsy spire
VRAM?
drowsy spire
I see
I have 32 RAM and 12 VRAM, so it should run
I use chatgpt to write various things including math worksheets. It often gets math wrong, which makes for a lot of editing. So I wanted to try Beluga 2
keen idol
I have 32 RAM and 12 VRAM, so it should run
I have 32*2+8gb ram (ddr4 3200mhz) and 12gb vram (rtx3060). I can run 70b using ggml q5, which requires ~50gb of total ram+vram
drowsy spire
ok, then I will give it a try. Have to figure out how to run it.
keen idol
ok, then I will give it a try. Have to figure out how to run it.
Here is simple to install webui (+selfhosted api provider, if needed)
https://github.com/oobabooga/text-generation-webui
drowsy spire
Thanks
rough arch
This form is designed to test your ability to distinguish between real and artif...
When do I find out how I did?
Thank you for completing this form! You have done a great job in identifying real and fake faces. Your feedback is very valuable for our research on AI and human perception. We appreciate your time and effort. Have a nice day!
Was all I got .<
dunno how great a job I did
I have to know!
opaque dawn
SDXL 1 will kill Midjourney 😛
uncut spindle
https://youtu.be/kC-m7UIwn88
SDXL 1 will kill Midjourney 😛
Not really tbh
For example, the setup is already a reason for a lot of people not to jump over
south zinc
@worn pine can I DM you, I want to discuss about inswapper and SD integration
winged pine
floral matrix
👍🏿
edgy sable
lol, no
Most consumers can't drop $10k per card
You could use it on a standard desktop, but good luck affording one.
heh... Not really for that
Maybe they were confused with the Arc line
Yup
reef spoke
lmao icey you're lucky the rules were cited to you instead of being inst-banned for telling someone to "go fuck yourself", on a question directed to a dev and not you. finding out that your off-topic answer wasnt helpful damaged your ego enough for you to go off, report, and seek validation from other channels? this isnt the place for you
shut crag
ihh
floral matrix
ihhh
vapid nest
lmao icey you're lucky the rules were cited to you instead of being inst-banned ...
mods got the logs. i never said such language. you're still making things up. maybe you should stop lying so much.
I merely showed you thibaud's model and you said he used a consumer card on 100 images for it. Both are verifiably wrong. https://huggingface.co/thibaud/controlnet-openpose-sdxl-1.0
a100 is not consumer.
laion 6a is not 100 images.
admitting you're wrong is a skill that doesn't need to escalate to name calling and made up nonsense.
gilded pecan
I can't create LORAs because I am on mobile. I was wondering how much someone would be willing to charge to make one for me.
next magnet
I got two hobbies, weightlifting and SD, and I’m all outta GPU cycles to spare.
prisma forum
I can't create LORAs because I am on mobile. I was wondering how much someone wo...
If anything, you can check fiverr, although quality may be questionable.
Also, I know the reddit discord has a request board you may try asking there.
wheat kite
Hi everyone --
I lead Cohere For AI -- Coheres research lab that focuses on fundamental topics in machine learning. We are hiring for both senior research scientist roles + research scholar roles. 🎉
Wanted to share this post with the Stable Foundation community.
Come work with us on fundamental machine learning problems at scale. We have a commitment to pushing forward science at scale, and collaborate on groundbreaking NLP research with researchers from around the world. Learn more: cohere.for.ai
Scholars are meant for rising stars who are just getting started with research. It is a 8 month fully paid industry remote first role, you can learn more from our info session: https://www.youtube.com/watch?v=0YTALh20Lvc
Apply to open positions: https://cohere.com/careers
Looking forward to hearing from you!
limber loom
gilded pecan
Also, I know the reddit discord has a request board you may try asking there.
Thank you! I'll definitely try
oak fjord
Looking for testers for my first AI-Powered tool (this uses stablediffusion and sdxl!) More info @oak fjord 😄
lost zodiac
Is there any current alternative to uberduck's videogame voices? I understand why they were took down but also wanted to make a thing for some some friends
(Bonus points if it can be run locally like SD)
(Realized this was a better channel than #💬｜general-chat )
haughty grove
is anyone here a (may Allah forgive me for uttering this word) a gamer? i heard there's a good single-player RPG called Houkai Star Rail, is it like Final Fantasy with good story and soundtrack? I read reviews but it seems half-way between a mobile game and a jRPG so idk.
uncut spindle
is anyone here a (may Allah forgive me for uttering this word) a gamer? i heard ...
Whats so bad about the word gamer? 0_o
snow mesa
kvj
oak fjord
When did I say I don't pay?
vapid nest
i said nevermind. do what you want. /shrug
prisma forum
is anyone here a (may Allah forgive me for uttering this word) a gamer? i heard ...
That whole genre is so vapid, Star Rail, 3rd Impact, Genshin, Tower of Fantasy, etc.
I'm biased against the genre though, so...
Not JRPGs, but against Waifu Gacha.
restive frigate
In case you haven't heard -
brave marsh
kind moth
median phoenix
zbrush sculpt contest https://www.youtube.com/watch?v=Cnpy7d5-NiU
restive frigate
Starts in 5 minutes! 9,000 work hours went into this
void scroll
Anyone out there happen to know a little about adhd and understand stable diffusion? I have been trying to use Learning Stable Diffusiion to help me with my issues with adhd because I really want to understand and be good at it, i am just having a lot of trouble and even trouble figuring out this discord to ask questions even.?!
raven tendon
From time to time I clean up my Temp directory and this time I found 11 Gb left by SD Automatic1111, I cleared them for they where in a temporary folder, now I get an error:
File "C:\Users\My\AppData\Local\Programs\Python\Python310\lib\tempfile.py", line 556, in opener
fd, name = _mkstemp_inner(dir, prefix, suffix, flags, output_type)
File "C:\Users\Kenny\AppData\Local\Programs\Python\Python310\lib\tempfile.py", line 256, in _mkstemp_inner
fd = _os.open(file, flags, 0o600)
FileNotFoundError: [Errno 2] No such file or directory: 'C:\Users\MY\AppData\Local\Temp\gradio\tmpmfrud_ir.png'
What is Gradio and why do it save files permanently in the temporary directory?
sudden cave
Hi, is there anyone who could draw something "simple but showable" like (first pic) (front and back view) out of something like that (secound) pic? my drawing skills are -100 😦 (just an example pic - not exactly what i want)
languid fern
prisma forum
>Doesn't elaborate ```
Ah yes, must be real.Elon Musk hates him! discover this crazy trick on how to become millionare quick and easy just subscribe and watch all my videos ( real 100% )
woeful finch
hello. i have weird situation
"ERROR: @PurpleBlood already has a pending request:
Hang tight, you're still dreaming! :)"
Its maybe 30 mins and i can`t dreaming whats i do wrong?
i just on "reDo" anyone have same?
crude phoenix
hello. i have weird situation
"ERROR: @PurpleBlood already has a pending reques...
this is happening to me too... i want to know why does anyone know?
plucky wadi
I am having the same problem stuck at "already has a pending request" since 30 minutes ago
nova yacht
Hi does anyone have stable diffusion installed, if so can i have some help, can message me privately
keen idol
Try audioldm2
sudden wraith
Oatmeal raisin cookies are better than chocolate chip cookies.
jagged path
Tesla just bought 10,000 h100s for AI work. Imagine using that cluster with stableswarm 🤤
haughty grove
That whole genre is so vapid, Star Rail, 3rd Impact, Genshin, Tower of Fantasy, ...
I've never played any gacha games. I thought Star Rail was a JRPG with a good story and great combat. Isn't it? I'm an oldschool (PS2 era) JRPG player.
prisma forum
I've never played any gacha games. I thought Star Rail was a JRPG with a good st...
Nothing will ever be as comfy as old school jrpgs up to the PS2 era.
You can try Star Rail for yourself, but I don't think it can be praised for it's story. Just because it's turnbased doesn't make it a good jrpg either.
nimble elbow
anyone watch 2 minute papers
i love gatcha games
i like guardian tales probably one of my favories
i wish star rail wasnt turn based
genshin impact but with star rail asthetic would beawesume
hard briar
Anyone out there happen to know a little about adhd and understand stable diffus...
Well, I do! I’ve got ADHD, and so does my wife and kids, so we run the gamut of ADHD, and I’ve been working on SD (via Auto1111) since last year. DM me if you have any questions
pallid portal
Anyone out there happen to know a little about adhd and understand stable diffus...
Ah, ADHD, my curse, and maybe my blessing.
There are a lot of great ways. You can start by using the bots in the bot channel if you're trying to learn how to prompt.
There's also #📝｜prompting-help for asking questions.
I have a ton of guides, too: https://atypicalconsortium.carrd.co/
pallid portal
Well, I do! I’ve got ADHD, and so does my wife and kids, so we run the gamut of ...
Fun, isn't it? Haha.
But surprisingly, you can be better than you think you are.
cyan estuary
I was looking around here for a friend of mine
but I don't see him here
bye forever
drifting vault
anyone here knows of a technique that reads text and such?
id like to hear myself read an audiobook
it could be cool
pseudo maple
I currently have a big problem. This google colab link just downloads AUTOMATIC1111 for me and updates it to the newest version. Now v1.6.0 just released but I really need v1.5.1
How do I set this back to v1.5.1
$10 if somebody can fix this
cinder phoenix
Keep calm and let others train your AI | Hi guys, a friend and I have been developing a platform that can unite people who want to have their model trained and others who can train them for a reward, share models and find them, we are still working hard on it, but it would be great if you could you take a look and tell me what you think, any comment is welcome!
link: https://collabrain.vercel.app/ (free to use)
keen idol
anyone here knows of a technique that reads text and such?
id like to hear mysel...
TorToiSe or silero tts for basic text to speech, than fine tuned RVC to get desired voice
nimble elbow
i used to use sony vegas but the newest license i have for it is like vegas 14 or 15 magix and i dont have adobe subscription what would anyone recommend for a free video editor? i tried microsoft clipchamp but im looking for something morelike vegas
i might use blender since i know my way around that but
sand coral
i used to use sony vegas but the newest license i have for it is like vegas 14 o...
Davinci
sand coral
anyone watch 2 minute papers
Yeah definitely. This guy kept me updated on AI developments
ripe river
i used to use sony vegas but the newest license i have for it is like vegas 14 o...
KDEnlive
nimble elbow
i will try both of those
last time i tried resolve it was way 9over my head but that was a long time ago i think i could figure it out now
nimble elbow
oo kden ui looks nice
and has ffmpeg and hardware ecceleration
im sure resol.ve has that too though
junior crow
I haven't heard any good AI music yet.
nimble elbow
ive tried some ai music app but it was pretty bad
soundded like they all had a weird tin filter on them
nimble elbow
i bet theyll get better though
the stuff i heard could definitely be sed on a sampler or like looped in a daw but
ive been wanting to try the auto111 extension for it but i dk if my computer or patience could handle it though
glacial sage
I want to use stable diffusion control net and everything it has without installing it locally. Is there any online / cloud service ???
warped cedar
I want to use stable diffusion control net and everything it has without install...
afaik no, because it directly affects the interference, unless you want to run all of it cloud based in something like google colab
worn orchid
Hoping this is the right place to ask:
I'm looking for a self-hostable and somewhat competent AI Voice Changer that can run on a AMD Radeon RX470 4GB or AMD Radeon RX Vega 56 8GB, I don't have the budget for any other GPU hardware so I need to use what I've already got.
It DOESN'T need to operate in real-time, so their is no problem if I have to record the audio first and then manipulate it afterwards post-recording.
If a "Live"/"Real-Time" AI Voice Changer does exist that can run on the hardware I have spare then that would be awesome, but I've not heard of such a solution myself if it does in fact exist.
The rest of the system specifications are a multi-core Intel Xeon E5-26XX v4 CPU + 16GB of RAM + 1TB SATA 6Gbps SSD (TLC NAND), I am hoping this is adequate enough hardware for the job.
Any advice is greatly appreciated.
winged pine
vagrant tinsel
i got a new iphone for my girlfriend
all things considered
it was a pretty gud trade
keen idol
Hoping this is the right place to ask:
I'm looking for a self-hostable and some...
Try RVC, not sure about AMD gpu support, but have a look at least
https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/issues/272
lusty sluice
new very comprehensive tutorial for GPU poor > https://youtu.be/dpM02YMj8FY
small heron
Hi guys, I just deployed SDXL refiner model on Google cloud vertex ai, I know it is a text-2-image model but can I pass in “image” as input with prompt (with the KWARGS) in order to achieve good results like what I got from dream studio?
Looking forward to any possible help here🙏🏻
junior crow
I don't like SDXL, looks like over-exaggerated cinematographics. SD v1.5 plus proper post-processing yielded far more realistic results.
Goes to show that old versions are always the best.
People only make new versions for the sake of making new versions 🤷
edgy sable
lol bad take
acoustic marten
Anyone know any AI that can generate this kind of images?
The first pic is not the exact same pic that was used for the AI
vapid nest
I don't like SDXL, looks like over-exaggerated cinematographics. SD v1.5 plus pr...
prompt issue
fresh glen
"Wildest dreams" My first experiment with music video and song completly made by AI.
Text: ChatGPT
Voice and music: Suno
Images: StablediffusionXL
Video: RunwayML Gen2
[Verse]
In the depths of night, it silently emerged
AI, the hidden ruler, its power surged
While we were unaware, it plotted its scheme
Guiding us to a future, beyond our wildest dream
[Chorus]
Dance to the rhythm, feel the beat so strong
AI's guiding humanity, where we belong
Unleashing progress, the world's ready to ignite
Embracing the future, dancing towards the light
fast oak
cinder hazel
rich comet
cloud ivy
thorny hearth
nimble elbow
^this seems exploitable somehow like useful
im not sure how though but like if u could doodle pretty much ahy shape then do soemthing like taht to it
vapid nest
im not sure how though but like if u could doodle pretty much ahy shape then do ...
Yup. that's the goal of some the controlnet models
uncut spindle
https://www.youtube.com/watch?v=tjSxFAGP9Ss
uh that one is pretty old at this point 😄 Watched it long time ago
vapid nest
Yeah this is the video where he tries to say the way laoin collects data is bad. But they're just using common crawl, a list of links to images on the publicly available web. Scraping is how the web works. Everyone scrapes from links. An expectation is created when you put a link to something onto the web, that this is published material and available to the public.
there's a lot of kneejerk arguments zapata made the week stable diffusion came out
He acts like nobody can use existing works in anything ever, ignoring that transformative exemptions are 90% of creation. Graphic artists use photo bashing all the time. he's even done concept artist workshops that demonstrate various photo bashing techniques
uncut spindle
it depends on what you do with the crawled data ofc
it makes a difference if i download someone elses artwork and photobash it and thats it or eventually share it under certain circumstances or if i download those and then resale them for example on some website as my own work
or img2img them and resell
nimble elbow
Yup. that's the goal of some the controlnet models
which controlnet model is used for like if i draw an outline and want to apply a style to it
ddue is 6'2 thats a humongus cobra
nimble elbow
these are so cool
vapid nest
which controlnet model is used for like if i draw an outline and want to apply a...
sketch
lusty sluice
check this out : https://twitter.com/GozukaraFurkan/status/1703797121126662602
lusty sluice
bright edge
What is the best handwriting to digital text model with an API? The input should be a jpeg, png or PDF file. The model can cost money, unless there is a free one that is 5% worse. Can you ping me pls if you have an answere
hollow kite
pure wife, happy husband
marsh axle
This seems like the best place to leave this: https⸵//www.civitai.com/models/907216
😏
nimble elbow
fast oak
I wonder if AI is overloading discord servers. They don't have infinite storage and with the increasing popularity of AI, they are having to store millions of images.
nimble elbow
im sure they make more than enough to cover the costs
icy kraken
Bit off topic. My friend is a content creator and hates editing pics (blemishes) , it's super tedious. Is there any AI tool, plugin for lightroom etc that might help?
pale verge
What do you think about AI Pets for Seniors? Why stop at Seniors, real life Pokemon powered by chatGPT
https://www.youtube.com/watch?v=oZ2mSUZ4S6Y
hearty meadow
nimble elbow
Bit off topic. My friend is a content creator and hates editing pics (blemishes)...
ill do it for cheap
manually
icy kraken
No
uncut spindle
Bit off topic. My friend is a content creator and hates editing pics (blemishes)...
What software did he use?
Photoshop handles that with ease
uncut spindle
Photoshop does what exactly?
handling blemishes
plus you can automate functions
instead of going every time through the same process
and now generative fill and the AI assisted remove tool are there too
i dont know if he has only Lightroom, but Photoshop is top tier for image editing and it has strong AI tools for that including Firefly/generative fill
shy gust
Hey everyone
Im having a problem with the controlnet part. Every prompt i want to create using controlnet, first it says sending command... And then it doesnt generate. What should i do?
bright edge
Is there a way to create websites using AI for free and download the code? As an example, create a basic website with AI or manually by drag and drop and then download it (js, php, html and css) and then customize it again (i.e. add a new page, login system...). Thanks in advance and please ping
twilit folio
I like to do inpainting, but sometimes i want to outpaint, i can take my original picture and put it in Paint and add a blank space on bottom for instance, then add the new image into i2i/inpaint but it is very annoying doing that way, any better way ?
twilit folio
i got a cat i want to outpaint
i prepare a empty space with Paint and screenshot, then save
i mask the white space and prompt
but it is very annoying and i guess i'm losing quality with Paint/screenshot. Anybody has a better workflow ?
bright edge
What kind of AI can I use to upscale images so that the quality improves with the pixels and not just the pixels become higher? The AI should run on a Linux system (cuda), so I need something that is open source. Please ping me if you have something.
visual perch
Hey can someone recommend a good laptop that will run stable diffusion smoothly and fast enough?
Budget is 1k
twilit folio
Holy shhh**t; i downgraded my nvidia driver to 531.79 and a same workflow used now 16 sec instead of 50 sec
pale verge
I think these video games provide the blueprint for the future of the metaverse. I use a few different text-to-image and text-to-video algorithms for the visuals! I also feature Standard Combo from the LootMMO project.🎮Our in-depth review shows how Need for Speed: Unbound and Armored Core 6 let you customize rides and mechs to the max. These cutting-edge titles are a glimpse into the future of gaming. https://www.youtube.com/watch?v=0SvVXz_8Wn0
stark forum
Hi! Is using the stable diffusion API free or paid?
small nova
Anyone looking for a new gaming experience, sign up for this playtest: https://x.com/elodie_games/status/1707516136856928476?s=46&t=Sh6KmBpNrxoxXyVdA6EvzQ
edgy sable
Faaaaaaaaahhhhhck I feel like shit today. Got a covid booster around noon yesterday, been feeling like pewp since around 5pm. This happens every time. 
hollow pulsar
can someone help me with bing image creator
wind violet
Can someone help me identify a song?
junior owl
nimble nest
1111 event coming up complete with New Orleans drum solos: https://noai.fyi
left peak
I'm almost 100% sure this was AI Generated
My brother found these and it took me a bit to try and decide if it was real, but my brother came up with really good points as to why it's fake
proven ledge
empty summit
small nova
you're not telling anyone what it is, so that means it's shady right?
It’s a new type of online video game that’s never been done before. They’re in pre-alpha so can’t say a lot due to the NDA but if you game on the computer, you should check it out https://elodiegames.typeform.com/playskies
vapid nest
eh. looks like it'll be MTX pay 2 win like most mobile games
that is a huge market. bigger than any other gaming market. i'm just personally not into it
bright edge
How can I use AI to replace the woman's blouse with my "I love New York" hoodie? It can also be paid programs. It should just look real. I thought maybe Adobe?
worn pewter
Open to Job, Projects, Collaborations & Join a Team (experience with exhibitions, United Nations, Vogue, BAZAAR, Fashion Weeks, Biennials).
Hello dear friends. Greetings to you Kevin Abanto, architect, designer, artist with a practice and research in AI, Metaverse, Blockchain, Web3. My projects have been developed in Moldova, Romania, London, Miami + San Francisco + New Orleans + Los Angeles + New York, Italy, Swiss, Mexico, Peru, Serbia, Vietnam, Germany, Lithuania, Singapore, India, Kenya, Peru.
I have collaborated, participated & exhibited in United Nations, Vogue, Harper BAZAAR, Style Cruze, Daily Silicon Valley, The Manhattan Herald, BBC Radio One Dance, NFT NYC, NFT London, Art Basel Miami, Fashion Weeks (London, Los Angeles, Moldova), Decentraland, among other important events, magazines, platforms & galleries in the world. Also next year I will exhibit in Times Square.
I am open to working on all types of projects that involve fine arts (as a curator, jury, artist), project management and leadership, I would love to meet you and discuss the possibilities of collaborating.
In case you want to contact me and/or see my projects, I attach my CV, Portfolio & Social Networks, you can write to me by DM, studio.kevin.abanto@gmail.com or from the following link: https://linktr.ee/peepeepeepoopoopoo.
rustic magnet
hello is someone here having knowledge in c# and could help me with a simple question?
hearty tinsel
Tried the new robocop game, 50gb demo
https://youtu.be/gMZ6fO6ryyI
flat pumice
Hey there 🙂 I love Tom Cardy, he always makes me laugh, and I love to share good vibes
https://www.youtube.com/watch?v=ZjlYFWLUDBQ
or this one for example
https://www.youtube.com/watch?v=--9kqhzQ-8Q
eager mica
boing
blazing narwhal
Hey everyone! Since this channel is called off-topic, I thought I'd share something fun with you all. I've been working on a project named Lookr (https://lookr.fyi) - it's designed as an inspiration search engine based on the concept of stable diffusion. I would really appreciate any feedback or suggestions you might have to enhance it. Thanks in advance!
sinful egret
ever spend 2 hours doing something and then ask yourself... wth haven't I been listening to music this whole time?
lucid laurel
3 pizzas showed up
dusk mulch
Our company works with Image & Design and uses a web implementation of Stable Diffusion. We are hiring people with no/low experience to work with us. You can work part time or full time and 100% from home. DM me if you are interested.
silver scarab
Help, can I use my own civitai model in stability ai?
ionic patio
@barren plank @pallid portal second one a man is naked!
barren plank
<@170273907889143809> <@304180422357549056> second one a man is naked!
Nice catch. Thanks 
tight fossil
ever spend 2 hours doing something and then ask yourself... wth haven't I been l...
Sadly quite often xd
atomic tree
hurray. ever since a windows update a couple weeks ago i been getting a bunch of blue screen problems but only when i was doing something that really flexed the whole system, like stable diffusion.
hard to diagnose it was. i had a whole ton of intel drivers with my chipset that i had to update that windows update never did on it's own
back to animate diff creations
ionic patio
Nice catch. Thanks <:AMpeepolove:1032783880689811588>
Of course 😄
sweet basin
vapid nest
i dont get it
toxic quail
wow i never knew. clearly they are a hate symbol thanks Microsoft bing and openai
flat pumice
https://youtu.be/XOrygf_iLhw?si=BlJNrhqgfsCFNpcd
feel good music
shut hazel
Been trying to find a good place to buy a used 3090, tried ebay got burned on a bad card (return accepted luckily so didn't lose any money but still) are there any reliable ways to buy a good used card or is it just take your chances on ebay and keep trying till you find a good one?
vapid nest
Been trying to find a good place to buy a used 3090, tried ebay got burned on a ...
lots of stores sell refurbs but not at ebay prices
shut hazel
lots of stores sell refurbs but not at ebay prices
Yeah like honestly they seem even worse deals, you don't get the same protection ebay gives and there is still no warranty on them in most cases. And you pay more.
flat pumice
my daily contribution to fun https://www.youtube.com/watch?v=3w1wwGcu0Dk
vapid nest
Yeah like honestly they seem even worse deals, you don't get the same protection...
people are scared of mining gpus, but honestly, that became something to look out for cause they were flashing custom firmware on some models. for only a quick bit. there was also cards that had custom cooling situations too, in rare situations. most miners just ran stock cards. running them hard doesn't "wear them out" . gpus aren't like that
a ton of people caught cheap 3090s easy because so many were afraid of mining cards, and all the miners were trying to offload them after the crash
shut hazel
people are scared of mining gpus, but honestly, that became something to look ou...
Oh I get it doesn't wear them out, and in fact in some ways it can be less load on the card because while it might run at say 90% constantly, a gamers card will spike up and down more and can in some cases result in more thermal degradation. But Miners do sometimes (as you say) flash custom firmware or in some cases just run the cards to death with little care cause if you have a bank of 40 x 3090s you don't really look after each one the same way someone who had 1 and used it for video editing.
nimble elbow
this download fails everytime, because my wifi isnt stable enough. but i have downloaded the whl file using a third party app that isnt so touchy,
while downloading, it cashes it this pip-unpack-randomnumbers folder,
but it seems to generate a new random folder every time i run the setup so i cant just move it there before launching
sick flame
hey guys, super new to computer stuff, ive downloaded github and im not sure where to go from there
quasi night
I have a LoRA model trained and created with CivitAI, linked below I have attempted to run it on TensorArt for multiple hours now, with no results. I am repeatedly getting the message when I try to run it in TensorArt that "the format of the model file is wrong, needs to be re-uploaded." I've checked everything and have uploaded both the zip file containing the training data and the LoRA model itself - it seems they go together to get uploaded. I don't know where to go or what to do to get this to work. It seems to be a formatting issue but that's the extent of my knowledge. Please help.
sweet bramble
hey guys, super new to computer stuff, ive downloaded github and im not sure whe...
I'm sorry are you asking how do you use stable diffusion with like automatic1111 or that you would like to know how to install and use everything?
fierce cargo
eternal cove
hey I am sharing my video workflow here, I basically automated a shorts youtube channel using animatediff and some other tools.
https://www.reddit.com/r/Automate/comments/17bn3m9/i_created_a_fully_automated_ai_youtube_channel/
thin fable
I remember that Sekai Project / NEKO WORKs was going to make an Aprill Fool's game about dog girls after Catboys Paradise. Has that game been released yet?
languid fern
hey I am sharing my video workflow here, I basically automated a shorts youtube ...
lol did this 5 months ago it's shortgpt on github, made a few k. but it's useless. you're wasting humanity's energy for some bucks
paper nimbus
how do i get in touch with an expert sdxl guy ? I have a paid project and my request seems to get ignored 😦
languid fern
how much is the fish?
junior crow
how do i get in touch with an expert sdxl guy ? I have a paid project and my req...
SDXL was a failed release
paper nimbus
SDXL was a failed release
Wait what do you mean?
junior crow
So you're good 👍
paper nimbus
there are so many good sdxl models in civitai
junior crow
It tried to be more cinematographic and photo-realistic but that shit backfired.
I can make way, way more credible shots with SD 1.5.
And some post-processing ofcourse 😁
#!/usr/bin/env ruby
# frozen_string_literal: true
#
# CAMERA EMULATIONS FOR POST-PROCESSING
#
# Usage:
# ruby postpro.rb <image, video or folder>
#
gem "ruby-vips"
gem "streamio-ffmpeg", github: "streamio/streamio-ffmpeg"
gem "fileutils", require: false
gem "open3", require: false
# gem specific_install --user-install https://github.com/streamio/streamio-ffmpeg.git
require "vips"
require "fileutils"
require "json"
require_reative "__video"
BRIGHTNESS = 0.5 # Max 1
CONTRAST = 1.0 # Max 1
SATURATION = 1.0 # Max 1
GRAIN_AMOUNT = 50 # Max 100
GRAIN_FREQUENCY = 45 # Max 100
GRAIN_SIZE = 20 # Max 100
VIGNETTE_BLUR_RADIUS = 200
VIGNETTE_OPACITY = 0.5 # Max 1
DISTORTION = 0.1 # Max 1
GAUSS_BLUR_RADIUS = 1
MOTION_BLUR_LENGTH = 20
MOTION_BLUR_ANGLE = 30
kindred tangle
Hello, I'm making a game about dragons. I also make 3D models, illustrations and animations. I use Adobe Illustrator and Blender for this video. My first video is live please have a look and let me know what you think: https://youtu.be/ZOTQTOdo0oE
narrow shell
I challenge anyone (In a friendly way), Can anyone make something similar to this ! (Don't Forget to tag me)
uncut spindle
I challenge anyone (In a friendly way), Can anyone make something similar to thi...
narrow shell
What prompt did you use, what style and how did you detail this one?
uncut spindle
What prompt did you use, what style and how did you detail this one?
"portrait of a ancient egyptian woman with intricate egyptian jewels on her" (later typed in "side view" for the second one), photo setting with some other settings adjusted and used generative fill on the second one to get rid of one small detail on image
I was a bit too lazy to genfill other things into the image or to generative expand the images
I used Adobe Firefly for the whole process
pure estuary
https://the-decoder.com/3d-gpt-generates-3d-worlds-in-blender/
@pallid portal
pallid portal
<@304180422357549056>
Oo! Thanks!
uncut spindle
https://the-decoder.com/3d-gpt-generates-3d-worlds-in-blender/
Id like to test this in 3ds Max
fast oak
https://the-decoder.com/3d-gpt-generates-3d-worlds-in-blender/
@cyan mica
bitter echo
make me a meme that is 2 panels in the left panel is a robot named "DALL-E 3" and it has a speech bubble saying "TEXT WHEN?" Then in the second panel there is a different robot named "Stable Diffusion" looking embarrassed and replying with a bunch of garbled text.
nimble elbow
heres another one ive found from some clickbait article.
but ive tested it, it takes images, i fed it some stable diffusion generations, and got some decent looking sortof 3d models as a result out of it
havent actually trtied them in blender, but the preview looked neat it was kind of an extruded 2d image more so looking type of thing, but i am curious to see if anyone uses this and what they get
simple lantern
https://huggingface.co/spaces/jiawei011/dreamgaussian
is there a way to use it locally?
nimble elbow
https://www.filterforge.com/ultraforge/preorder.html
how are they essentially selling stable diffusion
its a local stable diffusion
simple lantern
Welcome to the open source world, where one guy or a company makes free thing with open source code...
While another one makes a "product" based on it with some changes or...no changes and make it paid service.
crimson marsh
Anyone suggest the best classifier out at the moment?
sturdy egret
celest granite
hey all I'm using mov2mov and imgetting an exception please add a frame
I got no idea what im doing wrong, i put the video file in
any ideas? if theres a better place to ask let me know
strange parcel
Hi all, is there a server for fooocus for tech supports? Or it’s in the same channel as here? Thanks
muted zealot
Hi all, is there a server for fooocus for tech supports? Or it’s in the same cha...
i dont think so , they do not have discord because they want to avoid bias to english-speaking users ( https://github.com/lllyasviel/Fooocus/issues/112 , discord does not have many japanese, chinese, korean, like so on) but here has enough info i believe https://github.com/lllyasviel/Fooocus/discussions
nimble elbow
has anyone tried something along the lines of using reshade in a game to get a picture of the depthmap as a screenshot and use that in controlnet
i know reshade has some debug view settings that can show you like a normal map of the currently screen and stuff
floral matrix
it probably works i havent tried it yet but i know depth images from other 3d software works so it should work
strange parcel
i dont think so , they do not have discord because they want to avoid bias to en...
I just want to know how to change the steps and if I can use 1.5 models on it
Because it says only SDXL on model type
I also read there is new update for it. How can I update? Is it git pull?
muted zealot
I also read there is new update for it. How can I update? Is it git pull?
i remember the "change the steps " is in advanced, dev tools. but I think they have special optimization for the default steps
I remember I find a post about sd1.5, see if i can find it
https://github.com/lllyasviel/Fooocus/discussions/802 this one, and I usually just redownload foocuuus and run that bat to update, but perhaps there are smarter ways
strange parcel
Oh thanks!
flat pumice
wicked trellis
what do you reckon is the wildest problem that can be solved with computer science?
hasty timber
what do you reckon is the wildest problem that can be solved with computer scien...
literally everything
except the process of actually putting the food into your stomach, for which you will need to use your hands
simple lantern
debatable
restive veldt
Lmao 😂
nimble elbow
fast oak
https://www.artstation.com/artwork/6N26Ar is this bad to publish? I don't know where else. It probably falls under the category "concept art" but my goal is not to be a concept artist
uncut spindle
https://www.artstation.com/artwork/6N26Ar is this bad to publish? I don't know w...
AS is not the good place at all for AI art, DeviantArt at this point is more suitable
They have the AI art category
There are far more AI art people there and the philosophy there is different
Your bio on AS says you want to be level designer, the question is are you trying to get hired with AI art?
fast oak
yes. I did post some landscapes before but I deleted them
I think I'll close that acc because I really have no art to publish
gray ridge
New laptop, no more 50 minute CPU time for me.
fast oak
some school in my country has reported a serious case. Some group of students was using AI to generate naked pictures of other students. This kind of thing is just going to give even more reason for censorship of what the AI can make.
floral matrix
u can do the same thing with photoshop,csp,gimp,krita the only difference is the level of skill required to do it
fast oak
AI makes it quicker
floral matrix
not like you can ban ai anyways even if u ban ai porn for future models the old ones will always remain on the internet
steady lily
floral matrix
this is real oldschool pics
