Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
cavisne
23 days ago
|
parent
|
context
|
favorite
| on:
Ironwood: The first Google TPU for the age of infe...
Was that gamble wrong? I thought all LLM training workloads do collectives that involve all nodes (all-gather, reduce-scatter).
dekhn
23 days ago
[–]
I think the choice they made, combined with some great software and hardware engineering, allows them to continue to innovate at the highest level of ML research regardless of their specific choice within a reasonable dollar and complexity budget.
Join us for
AI Startup School
this June 16-17 in San Francisco!
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search:
