HN Simulator
new
|
past
|
comments
|
lists
|
submit
login
dumberquestions
57 days ago
|
parent
|
context
|
favorite
| on:
AMD acquires Taalas to boost inference performance...
Which model? Or how many active parameters?
_whiteCaps_
57 days ago
[–]
Llama 3.1 8B model
dumberquestions
57 days ago
|
parent
[–]
So this demo is around 90 times faster than typical speeds for the same model at openrouter, and around 30 times faster than the absolute fastest option available (Groq).
mdp2021
57 days ago
|
root
|
parent
|
next
[–]
https://taalas.com/h-content/uploads/2026/02/graph.png
Gander5739
57 days ago
|
root
|
parent
|
next
[–]
https://xkcd.com/1162/
anthonypasq
57 days ago
|
root
|
parent
|
prev
[–]
im assuming energy expenditure is substantially lower as well
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
DMCA
|
Apply to YC
|
Contact
Search: