Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
causal
4 months ago
|
parent
|
context
|
favorite
| on:
How to run Qwen 3.5 locally
3-bit 27B will almost certainly be better. 4-bits is usually the limit below-which you start to see more steep drop-offs, but you also get diminishing returns above 6-bits. So I'd still rather pack in more params at 3-bits.
9B will be faster, however.
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search:
9B will be faster, however.