“Leftists” being put over an image of the FUCKING POLICE

Is always funny.

OP has brain worms and is a lib

For simply productivity like Copilot or Text Gen like ChatGPT.

It absolutely is doable on a local GPU.

Source: I do it.

Sure I can’t do auto running simulations to find new drugs and protein sequencing or whatever. But it helps me code. It helps me digest software manuals. That’s honestly all I want

Also, massive compute projects for the @home project are good?

Local LLMs runs fine on a 5 year old GPU, a 3060 12 gig. I am getting performance on par with cloud ran models. I’m upgrading to a 5060ti just because I wanted to play with image Gen.

Whack. I just set up a Forgejo too.

Which is funny since that does solve a lot of the problems.

If it’s completely open source at least.

Like OS data sets and model that can be ran locally means it’s not trained on stolen data and it’s not spying on people for more data.

And if it runs locally on a GPU, it’s no worse for the environment than gaming. Really the big problem with the data center compute is the infrastructure of getting that data around.

I just assume anything that promises free stuff on Instagram reels is in fact a scam or a hack.

Also I’m not getting on that shit site to confirm it’s true.

15 minutes drive to my MIL and 4 hours to my own Mom.

My dad used to do tech support and wants to learn some of this stuff while he’s recovering from surgery and I’m at my MILs several times a month anyways. So it all works out. Also it’s only fair as the FIL has helped me do so much with my car over the years I wanted to pay them back and he likes movies more than me.

I’m a fan generally of LLMs for work, but only if you’re already an expert or well versed at all in whatever you’re doing with the model because it isn’t trust worthy.

If you’re using a model to code you better already know how that language works and how to debug it because the AI will just lie.

If you need it to make an SOP then you better already have an idea for what that operation looks like because it will just lie.

It speeds up the work process by instantly doing the tedious parts of jobs, but it’s worthless if you can’t verify the accuracy. And I’m worried people don’t care about the accuracy.

I download the YouTube tutorials I followed, upload them to my UAT Jellyfin server, and then when my server is having issues I can’t get to the videos!

A flawless system really