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

Stable Diffusion 2.0

stability.ai

Post reply on HN
Print view View original

61–70 of 519 posts

Re: Stable Diffusion 2.0

#61

Post by echelon »

kmeisthax wrote: post #58

Is there a good explanation of how to train this from scratch with a custom dataset[0]? I've been looking around the documentation on Huggingface, but all I could find was either how to train unconditional U-Nets[1], or how to use the pretrained Stable Diffusion model to process image prompts (which I already know how to do). Writing a training loop for CLIP manually wound up with me banging against all sorts of stra…

> train this from scratch

If you're talking about training from scratch and not fine tuning, that won't be cheap or easy to do. You need thousands upon thousands of dollars of GPU compute [1] and a gigantic data set.

I trained something nowhere near the scale of Stable Diffusion on Lambda Labs, and my bill was $14,000.

[1] Assuming you rent GPUs hourly, because buying the hardware outright will be prohibitively expensive.

Re: Stable Diffusion 2.0

#62

Post by dougmwne »

monero-xmr wrote: post #56
LASR wrote: post #27

I am a solo dev working on a creative content creation app to leverage the latest developments in AI. Demoing even the v1 of stable diffusion to the non-technical general users blows them away completely. Now that v2 is here, it’s clear we’re not able to keep pace in developing products to take advantage of it. The general public still is blown away by autosuggest in mobile OS keyboards. Very few really know how far…

There just isn’t a lot of market opportunities where being right 99% of the time is good enough. If you are operating at scale and 1/100 decisions are wrong, the outcome is poor and often highly off-putting to users. It’s possible this time is different, but people at my company were entertained by DALLE for all of 5 minutes before no one ever mentioned it again. The value proposition is simply low.

Are you kidding? Many times corporate decisions are being made effectively at random. Thinking that the average company operates with a 999 batting average is a total fantasy.

Re: Stable Diffusion 2.0

#63

Post by dougmwne »

bloaf wrote: post #50
rogers18445 wrote: post #3

They apparently tried to combat NSFW generation by filtering the training dataset not to include any.

They know they are going to be the next target in the war on general purpose computing. They're trying to stave it off for as long as possible by signalling to the authorities that they are the good guys. A confrontation is inevitable, though. Right now it costs moderate sums of money to do this level of training. Not always will this be so. If I were an AI-centric organization, I would be racing to position myself a…

I am not clicking that link because no one should take the risk of you proving your point of what horrors could pop out of one of these models.

I will say that while the government backlash is inevitable just like it was with encryption, these image generation models are so easy to train on consumer hardware that the cat is hopelessly out of the bag. It might as well be thoughtcrime.

Re: Stable Diffusion 2.0

#65

Post by pilaf »

88stacks wrote: post #60

Awesome, I’ve put stable diffusion on an api to train a model for anyone to use for free. I’m adding 2.0 to it as we speak! https://88stacks.com

> to sue for free

Typo of freudian slip?

Just kidding of course, nice project!

Re: Stable Diffusion 2.0

#66

Post by xnyan »

dougmwne wrote: post #62
monero-xmr wrote: post #56

Earlier quoted context omitted.

There just isn’t a lot of market opportunities where being right 99% of the time is good enough. If you are operating at scale and 1/100 decisions are wrong, the outcome is poor and often highly off-putting to users. It’s possible this time is different, but people at my company were entertained by DALLE for all of 5 minutes before no one ever mentioned it again. The value proposition is simply low.

Are you kidding? Many times corporate decisions are being made effectively at random. Thinking that the average company operates with a 999 batting average is a total fantasy.

When our c suite decides on an ad campaign and tells our artists to draw normal humans, those people have 3 legs or upside down teeth exactly 0% of the time. Humans have many many limitations, but with every model I’ve tested there’s a set of errors that would virtually never be made by any human.

Re: Stable Diffusion 2.0

#67

Post by jimbob45 »

bloaf wrote: post #50
rogers18445 wrote: post #3

They apparently tried to combat NSFW generation by filtering the training dataset not to include any.

They know they are going to be the next target in the war on general purpose computing. They're trying to stave it off for as long as possible by signalling to the authorities that they are the good guys. A confrontation is inevitable, though. Right now it costs moderate sums of money to do this level of training. Not always will this be so. If I were an AI-centric organization, I would be racing to position myself a…

Banknote printing is primarily protected against on the hardware level of printers, no? With the nigh-invisible unique watermark left by every printer, there’s virtually no way you’d get away with it. My guess is that the Photoshop filter exists mostly as a barrier against the crime of convenience.

Re: Stable Diffusion 2.0

#68

Post by joshspankit »

Gigachad wrote: post #11
argella wrote: post #6

Earlier quoted context omitted.

Bummer. AI porn is fun.

The future is probably models trained almost exclusively on porn.

Porn has driven many tech advances. I predict that models trained on specific porn genres will appear as soon as training a good model is doable for under $5000. They’ll get here much quicker if we get video to that mark first.

Re: Stable Diffusion 2.0

#70

Post by llui85 »

dougmwne wrote: post #63
bloaf wrote: post #50

Earlier quoted context omitted.

They know they are going to be the next target in the war on general purpose computing. They're trying to stave it off for as long as possible by signalling to the authorities that they are the good guys. A confrontation is inevitable, though. Right now it costs moderate sums of money to do this level of training. Not always will this be so. If I were an AI-centric organization, I would be racing to position myself a…

I am not clicking that link because no one should take the risk of you proving your point of what horrors could pop out of one of these models. I will say that while the government backlash is inevitable just like it was with encryption, these image generation models are so easy to train on consumer hardware that the cat is hopelessly out of the bag. It might as well be thoughtcrime.

Link doesn't show any model output - it's an screenshot of photoshop refusing to edit a banknote.
Post reply on HN