Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
Oras
5 months ago
|
parent
|
context
|
favorite
| on:
Qwen3.6-27B: Flagship-Level Coding in a 27B Dense ...
If these models reach quality of Opus 4.5, then DGX could be a good alternative for serious dev teams to run local models. It is not that expensive and has short time to make ROI
czk
5 months ago
[–]
Memory bandwidth is the biggest L on the dgx spark, it’s half my MacBook from 2023 and that’s the biggest tok/sec bottleneck
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search: