Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
LarsDu88
on Sept 18, 2024
|
parent
|
context
|
favorite
| on:
We are self-hosting our GPUs
Does each machine have just 1 gpu per rig or is it multiple? Do you network them to do DDP or is it just 1 inference job per card?
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search:
