Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
svara
23 days ago
|
parent
|
context
|
favorite
| on:
TPUs vs. GPUs and why Google is positioned to win ...
... but not in deep learning or am I missing something important here?
qeternity
22 days ago
[–]
Yes, absolutely in deep learning. Custom fused CUDA kernels everywhere.
Scene_Cast2
22 days ago
|
parent
[–]
Yep. MoE, FlashAttention, or sparse retrieval architectures for example.
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search:
