Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
cardine
on Jan 2, 2023
|
parent
|
context
|
favorite
| on:
Petals: Run 100B+ language models at home bit-torr...
A lot of computation is offloaded to the CPU, such as gradients and optimizer states. You are right though that quite a bit of computation is still done on the GPU.
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search: