GitHub
KVarN is a native vLLM KV-cache quantization backend for your agents: 3-5x more context, throughput above FP16, and FP16-level accuracy. Calibration-free, one flag.
On GitGem
Written in Python, licensed under Apache-2.0.
It has been tracked on GitGem.org since June 4, 2026. It has 420 stars and 27 forks, with its most recent commit 23 days ago.
See the Python trending boardStars
420
Forks
27
Open Issues
5
Language
Python
License
Apache-2.0
Last Commit
3w ago
Comments
No comments yet. Be the first!
You will sign in with GitHub to post. Your draft is kept.