News for hackers. Stuff that matters.
Re:Stable Diffusion 1.5 by lysp on Thursday October 20, 2022 @11:24PM (#33284123)
If you have a reasonable video card, you can run it easily locally using this repo:

https://github.com/AUTOMATIC1111/stable-diffusion-webui/

It is extremely active - author updates it 10-20 times per day.

Re:Stable Diffusion 1.5 by ronsor on Friday October 21, 2022 @12:28AM (#33284505)
IMO the local UIs are still designed like a slot machine :)
Re:Stable Diffusion 1.5 by 13of40 on Friday October 21, 2022 @12:34AM (#33284548)
TBH, I bought a 3090 and it broke after a week so I sent it back. It was about $1200, and I'm down about $80 on the web site at this point, so... I have been pondering a cloud solution though, but haven't done the math all the way yet. It's a bit like buying a $15K boat and then just going out on it once a month.
Re:Stable Diffusion 1.5 by danuker on Friday October 21, 2022 @04:43AM (#33285944)
> and it broke after a week

Have you monitored its temperature while using it? Did your warranty cover it?

Re:Stable Diffusion 1.5 by imhoguy on Friday October 21, 2022 @02:41AM (#33285211)
SD even runs fine, although slower, on MacBook Air M1 thanks to quite capable GPU which shares RAM as VRAM, I think soon to utilize Neural Engine Cores too. Announcement and discussion: https://www.reddit.com/r/StableDiffusion/comments/xbo3y7/one...
Re:Stable Diffusion 1.5 by Huh1337 on Friday October 21, 2022 @06:45AM (#33286616)
Slower means around a minute per picture. It's amazing! Guys with GPUs more expensive than my whole computer can't run it, but a laptop without fans can. Well done Apple.
Re:Stable Diffusion 1.5 by polyanos on Friday October 21, 2022 @02:28PM (#33291400)
I have a old ass GTX1070 and run it just fine, it's not the fastest but it works. A lot of these cases are just PEBCAK's and should just be removed from the internet, it's been made so easy that a literal child could install and use SD.

Well as long as they have a recent-ish NVidia card, rip AMD users.

Re:Stable Diffusion 1.5 by 13of40 on Friday October 21, 2022 @03:49PM (#33292314)
The 3090 I had was actually somewhat slower than dreamstudio.ai, but the nice thing was the set of python tools I got with it was scriptable so I could do things like create an image, use that as input for another one, and so on, then make them into an animation. There's a bit of tax with initializing python and loading the ML model with each iteration, but if I ever get my hands on a 4090 I'm sure I can solve that.