Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
pjm331
15 days ago
|
parent
|
context
|
favorite
| on:
Qwen3.8-Max: A New Bar for Coding and Cowork
How’s that been going?
harhargange
11 days ago
[–]
The 27b model inference speed is amazing at 35 tok/s with power limit to 260W. My cpu is 5950x, maybe that matters too. Really enjoying it and it’s my default now.
reply
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search:
