Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
apitman
36 days ago
|
parent
|
context
|
favorite
| on:
Qwen 3.8 27B
I wonder how this would stack up against 4x RTX 3060, assuming you have the physical room for them.
Almondsetat
36 days ago
[–]
Since the 3060s would have to communicate constantly using pcie, I think I'd go with intel
apitman
35 days ago
|
parent
[–]
Currently software is king when it comes to inference performance though.
Almondsetat
35 days ago
|
root
|
parent
[–]
Having the entire model in the vram without having to transfer anything is even more king, though
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search: