Settings

Theme

KVarN: Native vLLM backend for KV-cache quantization by Huawei

github.com

49 points by theanonymousone 2 hours ago · 8 comments

Reader

throwa356262 an hour ago

Better performance than TQ and better quality than FP16?

Am I reading this right??

v3ss0n an hour ago

Why this is not a PR for vLLM ?

  • esafak an hour ago

    It's the output of a research paper; the authors are not trying to build up vLLM, and they probably have no incentive to do so. You can submit a PR, though! It's easier now while the divergence is low, so don't wait. Since there are six authors, I bet you could get help with the inevitable review chores if you just take the step of creating the PR.

    edit: It might not be clear that it is based on vLLM 0.22, which is the current version: https://github.com/huawei-csl/KVarN/commit/d6290e99098d7426d.... All you have to do is create a diff off it; it's fairly straightforward.

    • jmalicki an hour ago

      And with the help of AI, pointing at AI at this paper and saying "making a vLLM PR from this paper" tends to work surprisingly well, even if you need to nudge it a little bit along the way.

Keyboard Shortcuts

j
Next item
k
Previous item
o / Enter
Open selected item
?
Show this help
Esc
Close modal / clear selection