Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
boulos
10 months ago
|
parent
|
context
|
favorite
| on:
Lossless LLM compression for efficient GPU inferen...
The latest GPUs and TPUs support fp8. It's a big part of the efficiency gain in the latest systems. Blackwell also supports fp4.
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search:
