Y
Hacker News
new
|
ask
|
show
|
jobs
rohansood15
10 hours ago
[
-
]
The paper is about vector quantization, which affects KV cache not model weights/sizes.
Please enable JavaScript to continue using this application.