this post was submitted on 25 Mar 2025
36 points (97.4% liked)

Technology

1097 readers
81 users here now

A tech news sub for communists

founded 2 years ago
MODERATORS
muad_dibber@lemmygrad.ml
36
DeepSeek-V3 now runs at 20 tokens per second on Mac Studio, and that’s a nightmare for OpenAI (venturebeat.com)
submitted 2 weeks ago by yogthos@lemmygrad.ml to c/technology@lemmygrad.ml
15 comments fedilink hide all child comments
you are viewing a single comment's thread
view the rest of the comments
[–] CriticalResist8@lemmygrad.ml 4 points 2 weeks ago (1 children)

But imagine what you'll be able to run it on in four more months. But yeah, it's stretching the definition of consumer hardware a bit.

[–] pinguinu@lemmygrad.ml 1 points 2 weeks ago (1 children)

You can use the smaller models on (beefy) consumer hardware already. That's something, right? 😅

[–] CriticalResist8@lemmygrad.ml 3 points 2 weeks ago (1 children)

I want the full 1TB model running on my 10 year old linux laptop

[–] pinguinu@lemmygrad.ml 2 points 2 weeks ago

Just put your persistent memory as swap. Easy