This new Chinese AI Is outperforming ChatGPT — and its open-source & runs locally

https://www.tomsguide.com/ai/this-new-chinese-ai-is-outperforming-chatgpt-and-it-runs-locally

53 points · 9 comments · view on lemmy.world

9 Comments

dimjim@sh.itjust.works · 34 pts · 43d (4 replies)

Yet, even when heavily compressed, it requires roughly 240GB of memory just to load.

Ah I'll just pop it in the ol' Raspberry Pi then, easy peasy.

Nouvellalia@lemmy.world · 6 pts · 43d

Lol, "runs locally". I mean, Claude rubs locally too if you're in the room with the racks.

Edit: I said what I said. Get some lube and go hang out with Claude's hot, noisy, 5kw rack. You know what they say "Once you go stack you never go back."

Valmond@lemmy.dbzer0.com · 5 pts · 43d

RPi 6: 256GB RAM

kutt@lemmy.world · 3 pts · 43d

Currently vibe coding with it running on my vape!

fluffykittycat@slrpnk.net · 2 pts · 43d

Basically they never had any moat to begin with but no one else seems to know how to fit that much intelligence into less space. It's possible that it just fundamentally has to take up that much space which would also imply that future Computing gainss are going to be more focused on memory than raw competition

pennomi@lemmy.world · 28 pts · 43d (2 replies)
[ removed ]
timewarp@lemmy.world · 21 pts · 43d

Sucks that people lump AI into a single category of whatever cloud-hosted subscription that tech bros from Silicon Valley are pushing.

fluffykittycat@slrpnk.net · 2 pts · 43d

Given how memory is the bottleneck especially at the very low end it makes me wonder if one bit quantization of an extremely large model would be a gigabyte per gigabyte of ram better

eleijeep@piefed.social · 25 pts · 43d

GLM 5.2

saved you a click

_haha_oh_wow_@sh.itjust.works · 5 pts · 43d

C'mon, pop!