paper-with-me

Papers

PackKV: Reducing KV Cache Memory Footprint through LLM-Aware Lossy Compression

2025-12-30 · Bo Jiang, Taolue Yang, Youyuan Liu, Xubin He, Sheng Di, Sian Jin arxiv

Transformer-based large language models (LLMs) have demonstrated remarkable potential across a wide range of practical applications. However, long-context inference remains a significant challenge due to the substantial memory requirements of the key-value (KV) cache, which can scale to several gigabytes as sequence length and batch size increase. In this paper, we present \textbf{PackKV}, a generic and efficient KV cache management framework optimized for long-context generation. %, which synergistically supports both latency-critical and throughput-critical inference scenarios. PackKV introduces novel lossy compression techniques specifically tailored to the characteristics of KV cache data, featuring a careful co-design of compression algorithms and system architecture. Our approach is compatible with the dynamically growing nature of the KV cache while preserving high computational efficiency. Experimental results show that, under the same and minimum accuracy drop as state-of-the-art quantization methods, PackKV achieves, on average, \textbf{153.2}\% higher memory reduction rate for the K cache and \textbf{179.6}\% for the V cache. Furthermore, PackKV delivers extremely high execution throughput, effectively eliminating decompression overhead and accelerating the matrix-vector multiplication operation. Specifically, PackKV achieves an average throughput improvement of \textbf{75.7}\% for K and \textbf{171.7}\% for V across A100 and RTX Pro 6000 GPUs, compared to cuBLAS matrix-vector multiplication kernels, while demanding less GPU memory bandwidth. Code available on https://github.com/BoJiang03/PackKV

📄 PDF Abstract BibTeX arXiv:2512.24449

Code (0)

등록된 구현이 없습니다.

Tasks

Computational Efficiency

Similar Papers 제목 키워드 기반

HeadInfer: Memory-Efficient LLM Inference by Head-wise Offloading

2025-02-18 · Cheng Luo, Zefan Cai, Hanshi Sun, Jinqi Xiao 외

Transformer-based large language models (LLMs) demonstrate impressive performance in long context generation. Extending the context length has disproportionately shifted the memory footprint of LLMs during inference to t…

Computational EfficiencyCPUGPU

KV-Compress: Paged KV-Cache Compression with Variable Compression Rates per Attention Head

2024-09-30 · Isaac Rehg

Context lengths of Large Language Models (LLMs) have exploded in recent years, with 128k-token context becoming a standard and million-token context becoming a reality. Efficiently supporting long-context inference remai…

Stochastic KV Routing: Enabling Adaptive Depth-Wise Cache Sharing

2026-04-03 · Anastasiia Filippova, David Grangier, Marco Cuturi, João Monteiro arxiv

Serving transformer language models with high throughput requires caching Key-Values (KVs) to avoid redundant computation during autoregressive generation. The memory footprint of KV caching is significant and heavily im…

MiniCache: KV Cache Compression in Depth Dimension for Large Language Models

2024-05-23 · Akide Liu, Jing Liu, Zizheng Pan, Yefei He 외

A critical approach for efficiently deploying computationally demanding large language models (LLMs) is Key-Value (KV) caching. The KV cache stores key-value states of previously generated tokens, significantly reducing …

Quantization

OrbitFlow: SLO-Aware Long-Context LLM Serving with Fine-Grained KV Cache Reconfiguration

2026-01-05 · Xinyue Ma, Heelim Hong, Taegeon Um, Jongseop Lee 외 arxiv

Serving long-context LLMs is challenging because request lengths and batch composition vary during token generation, causing the memory footprint to fluctuate significantly at runtime. Offloading KV caches to host memory…