SEP 10, 2026 · PREPRINT
Building py-kvcache: A Performance Characterization of External KV Caching for vLLM with NVMe SSDs
arXiv
This is a systems performance characterization study of an experimental software implementation (py-kvcache) for optimizing LLM inference, presented as a preprint without peer review; it demonstrates engineering improvements in a specific tool but lacks clinical or health outcomes and is not yet published in a peer-reviewed venue.
Reported
Speedup vs LMCache at 80k tokens2.0x
Preloading contribution1.34x
Speedup vs LMCache with GPU/CPU/d…1.23x