Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
lostmsu
on Dec 19, 2023
|
parent
|
context
|
favorite
| on:
AMD's CDNA 3 Compute Architecture
Microsoft has DirectML and a PyTorch backend for it. Although it is stale and significantly slower on 3090 than CUDA (5x+ as of 2 years ago).
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search:
