Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
ta988
59 days ago
|
parent
|
context
|
favorite
| on:
Apple M3 Ultra
You will have to explain to me how.
bmelton
59 days ago
|
next
[–]
https://digitalspaceport.com/how-to-run-deepseek-r1-671b-ful...
muricula
59 days ago
|
parent
|
next
[–]
Is that a CPU based inference build? Shouldn't you be able to get more performance out of the M3's GPU?
wmf
59 days ago
|
root
|
parent
|
next
[–]
Inference is about memory bandwidth and some CPUs have just as much bandwidth as a GPU.
johnmaguire
59 days ago
|
prev
[–]
https://news.ycombinator.com/item?id=42897205
Join us for
AI Startup School
this June 16-17 in San Francisco!
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search:
