|
|
|
|
|
|
by Iolaum
10 days ago
|
|
|
Strix Halo chips have around 210+ GB/sec gpu memory bandwidth and announcements put the new nvidia chip at around 300GB/sec gpu memory bandwidth. I 'd say that is an improvement if you want to run local llm inference. Still well below with what you can achieve with Apple chips though. |
|
|