The debate about local vs cloud AI models is driving me crazy

I spent 6 hours trying to get a local LLM to run on my laptop last weekend and it still crashed. Has anyone else found that cloud tools are way more reliable, or is running things locally worth the headache for privacy reasons?

3 comments

3 Comments

wadem897d agoMost Upvoted

Man, that's the exact same thing I went through. Tried getting Ollama working on my old gaming laptop and it just kept crashing every time I tried to load any model bigger than 7B. Cloud stuff just works right out of the box, no tweaking drivers or fighting with VRAM limits. But yeah, I do feel kind of weird sending my personal stuff to some server somewhere... still haven't figured out which way is better.

sean7827d ago

Honestly, I see it the other way. I ran LLaMA 3.1 8B on my work laptop without much trouble, and I've found the local stuff way more consistent once you get it set up. Cloud tools go down or change their pricing all the time, and I've lost access to stuff mid-project before. Plus, if you're doing anything even slightly sensitive - like helping a buddy with personal document analysis - sending that to a random server feels wrong. Took me maybe an hour to tweak the settings and get Ollama stable, now it just works.

bennett.mason7d ago

Took me a weekend of tinkering but once I got it dialed in, haven't touched cloud tools since.