cd/entity/KVarN· home› entities› KVarN
grep -l @kvarn /news/*.json | wc -l → 1

@KVarN

mentions 1 type Organization feed RSS
15:18
2026-06-04
github.com
large-language-models

KVarN: Native vLLM KV-cache quantization back end by Huawei

Huawei released KVarN, a native KV-cache quantization back end for vLLM that delivers up to 5x more cache capacity and 1.3x the throughput of FP16 while maintaining FP16-level accuracy. The calibratio…

// co-occurs with top 5 entities
Huawei 1 vLLM 1 TurboQuant 1 Qwen3-32B 1 AIME25 1