🛰️ Daily AI Frontier
‹ back to 2026-08-15

vToken: Token-Level Virtualization for Reclaimable KV Caches

Research Efficiency & Systems

Ranking

Overall 80
Content 100
Popularity 34

Observed public metrics from 1 member.

Merged summary

TL;DR - vToken virtualizes KV-cache management at token granularity, enabling serving systems to reclaim memory stranded within fixed-size blocks. Its vLLM implementation improves memory utilization, throughput, and concurrency without changing PagedAttention kernels.

  • Decouples logical token liveness from physical placement using token-table indirection.
  • Asynchronously repacks live tokens while preserving CUDA Graph compatibility.
  • Reduces retained KV blocks per request by 27.2%–72.3% and improves SLA-constrained throughput by up to 1.37×.
  • Supports up to 2× more concurrency under constrained active-KV budgets and reduces policy integration from 500+ lines to under 50.

Sources (1)

vToken: Token-Level Virtualization for Reclaimable KV Caches

arXiv cs.AI Yuanhang Gao, Xiangrui Yang, Yuanfeng Chen, Hongjia Chen, Qianru Lv, Wenfei Wu, Dongsheng Li 2026-08-13 arXiv:2608.13263
Public signals Semantic Scholar citations 0 · Semantic Scholar influential citations 0
Providers: Hugging Face · N/A OpenAlex · N/A Publisher · N/A Semantic Scholar · Citations 0 · Influential citations 0 X · N/A Fetched 2026-08-30 14:18:19.880092 UTC

TL;DR - vToken virtualizes KV-cache management at token granularity, enabling serving systems to reclaim memory stranded within fixed-size blocks. Its vLLM implementation improves memory utilization, throughput, and concurrency without changing PagedAttention kernels.

  • Decouples logical token liveness from physical placement using token-table indirection.
  • Asynchronously repacks live tokens while preserving CUDA Graph compatibility.
  • Reduces retained KV blocks per request by 27.2%–72.3% and improves SLA-constrained throughput by up to 1.37×.
  • Supports up to 2× more concurrency under constrained active-KV budgets and reduces policy integration from 500+ lines to under 50.
item →