I like LLMs but I want to run them locally and for that I need RAM that I can’t really afford. Especially Chinese models are good candidates to run locally, I wonder if that’s part of the US strategy, I mean making local inference so expensive that people have to use their models if they don’t want to send all their data to China
I think we can use sam altman to our advantage
Submitted 3 hours ago by cm0002@suppo.fi to memes@sopuli.xyz
Comments
affenlehrer@feddit.org 1 hour ago
Klear@piefed.world 2 hours ago
Yeah, like this trick wouldn’t work on any of us…
Sibbo@sopuli.xyz 2 hours ago
Infinite energy unlocked!
Diplomjodler3@lemmy.world 2 hours ago
Damnit, Sam! What are you doing under my desk?!?
cannedtuna@lemmy.world 2 hours ago
Put him on a treadmill to power a data center
deadbeef79000@lemmy.nz 2 hours ago
Just lead him to a deep hole and let him fall in.