Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Running Stable Diffusion XL 1.0 in 298MB of RAM

github.com

Post reply on HN
Print view View original

111–120 of 166 posts

Re: Running Stable Diffusion XL 1.0 in 298MB of RAM

#111

Post by pmarreck »

codetrotter wrote: post #83
pmarreck wrote: post #66

Earlier quoted context omitted.

This is more than a little melodramatic. https://frame.work/ and the https://mntre.com/ MNT Reform: Exist

If my country decides to ban the ownership of general purpose computers for individual persons, they would order the customs service to stop import of any computer hardware that enabled general purpose computing. Now I would not be able to have any computer shipped to me from outside my country, so I could no longer buy from either of those vendors you linked. Furthermore, it also would mean that I would not be able…

Banning the import of personal computers would be absolutely disastrous for any possible economy anywhere.

Re: Running Stable Diffusion XL 1.0 in 298MB of RAM

#112

Post by IKantRead »

Drakim wrote: post #47
samr71 wrote: post #9

So this should be it for trying to regulate stable diffusion type tech, right? If these models and their inference infra can be shrunk down to be runnable on a PS2, it doesn't seem like it's possible to stop this tech without a totalitarian surveillance state (and barely even then!).

The war on general computing has been ongoing but not made enough inroads to stop people from owning general computing devices (yet)

I try to bring up as often as possible in conversation that nearly all the progress we're seeing in terms of usability and performance is precisely because of the open source support for these models.

Especially because these tools are so popular outside of the developer community, I think it's worth really beating into peoples minds that without open source AI would be in a much worse place overall.

Re: Running Stable Diffusion XL 1.0 in 298MB of RAM

#113

Post by froggit »

maxlin wrote: post #38
SigmundurM wrote: post #7

Earlier quoted context omitted.

"0.5-2x slower" must be a typo on their part right? If something is 0.5x slower, then it is 2x faster. I assume they meant to say "1.5-2x slower".

Communication is hard. If they said "Takes 50% to 200% more time" it would have been clearer

What? No, that's confusing enough it's almost hostile. The fact thst your math is wrong is proof enough. 50% to 200% more time is 1.5x to 3x slower.

I don't like how it was worded by the author. But all you've done is essentially invert the wording while making the math MORE difficult in the process.

Re: Running Stable Diffusion XL 1.0 in 298MB of RAM

#114

Post by wayfinder »

froggit wrote: post #113
maxlin wrote: post #38

Earlier quoted context omitted.

Communication is hard. If they said "Takes 50% to 200% more time" it would have been clearer

What? No, that's confusing enough it's almost hostile. The fact thst your math is wrong is proof enough. 50% to 200% more time is 1.5x to 3x slower. I don't like how it was worded by the author. But all you've done is essentially invert the wording while making the math MORE difficult in the process.

Umm what? I hear “50% more time” all the time.

50% to 200% is 0.5x slower to 2x slower.

People seem to be confusing “% slower/more time” vs “% of current time”

Re: Running Stable Diffusion XL 1.0 in 298MB of RAM

#115

Post by mkaic »

riskable wrote: post #74
Nevermark wrote: post #61

Impressive! Verily, the era is nigh wherein even lamps and toasters shall brim with surpassing sagacity. After exposure to this field for many years, the last decade was stunning. I say “was”, because the speedup in the last 6-18 months has been another thing altogether. I am not concerned with what we will be able to two years hence, but with how much faster progress will be. And then again, and again.

Ooh! A toaster that takes a prompt and generates that image on your toast! The GPU heat could be harnessed to actually toast the toast. Let's make a startup!

We're extremely proud to announce that ToasterDream has raised $323M in Series A funds, and we look forward to many years of exciting developments ahead. As the CEO I'd like to personally assure our loyal customers that taking this funding will not compromise the quality of our goods and services — rather, quite the opposite! In fact, in just the first month since we secured this funding with our investors, we have already managed to create an entirely new product! It's called ToasterDream Ultra, and allows users to toast up to 8 images simultaneously for just $5.99/month...

Re: Running Stable Diffusion XL 1.0 in 298MB of RAM

#116

Post by BHSPitMonkey »

agilob wrote: post #18
amelius wrote: post #13

Earlier quoted context omitted.

Not very environment-friendly, though.

Dude, we are literally using single use plastic bottles to store water for a few weeks in them.

If we're doing one bad thing already, we may as well do a hundred!

Re: Running Stable Diffusion XL 1.0 in 298MB of RAM

#118

Post by lcnPylGDnU4H9OF »

emi2k01 wrote: post #60
croes wrote: post #52

Earlier quoted context omitted.

It depends. What do you think if I say it's 1x slower? Is it as fast as the original or does it take twice as long?

Twice as long. As fast as the original would be "0x slower" or "1x as fast". People should just use duration instead of speed as you did at the end: "takes twice as long", "takes 1/3 of the time..."

> As fast as the original would be "0x slower" or "1x as fast".

I have played a fair number of incremental games and this quickly became a pet peeve of mine. So many will say things like "2x more" and it will actually be "2x as much". Fortunately, I don't recall any which actually switch between the meanings but it's so commonly a guessing game until I figure it out.

Re: Running Stable Diffusion XL 1.0 in 298MB of RAM

#119

Post by ethbr1 »

tavavex wrote: post #77
ethbr1 wrote: post #71

Earlier quoted context omitted.

An extra-territorial crime Mostly committed by private citizens in pursuit of profit That all nations of the world have an interest in suppressing to encourage free trade that economically benefits them But which some countries at various times have a geopolitical interest in supporting ... you're right, they have no logical or legal connections at all.

You could tie essentially any two crimes by assigning more broad descriptors to them that'd boil down to "this is what countries want to discourage". Not to mention, half of this is just wrong - digital piracy most often isn't extraterritorial (it very much falls under the jurisdiction of where the piracy took place), and most individuals pirate for personal needs, not profit. The point stands - no jurisdiction that…

> digital piracy most often isn't extraterritorial (it very much falls under the jurisdiction of where the piracy took place)

The canonical eBay/PayPal fraud from eastern Europe example?

> most individuals pirate for personal needs, not profit.

But most piracy is done by individuals in pursuit of profit, not for personal need.

Re: Running Stable Diffusion XL 1.0 in 298MB of RAM

#120

Post by leonidasv »

refulgentis wrote: post #23
Filligree wrote: post #22

Earlier quoted context omitted.

Okay... what's the downside?

In terms of, what's the tradeoff for the time decrease? Apples to oranges, they're comparing 11 hours on a Raspberry Pi Zero to: - 10 seconds on Intel i7-13700 - 3 seconds on Intel i9-9990XE - 5 seconds on Ryzen 9-5900X Additionally, the 2048 is accomplished by using RealESRGAN to 2x, which isn't close to what a native 2048 diffuser's quality would be. It does look interesting and is an achievement, in terms of, it's…

Unrelated, but now I'm curious about how much would it take on RPis 4 and 5.
Post reply on HN