Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
xiphias2
11 months ago
|
parent
|
context
|
favorite
| on:
Qwen3: Think deeper, act faster
When I tested Qwen with different sizes / quants, generally the 8-bit quant versions had the best quality for the same speed.
4-bit was ,,fine'', but a smaller 8-bit version beat it in quality for the same speed
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search:
4-bit was ,,fine'', but a smaller 8-bit version beat it in quality for the same speed