KV Pareto: Systems-Level Optimization of KV Cache and Model Compression for Long Context Inference

• URL: http://arxiv.org/abs/2512.01953v1
• Date: Mon, 01 Dec 2025 18:03:47 GMT
• Title: KV Pareto: Systems-Level Optimization of KV Cache and Model Compression for Long Context Inference
• Authors: Sai Gokhale, Devleena Das, Rajeev Patwari, Ashish Sirasao, Elliott Delaye,
• Abstract summary: Long-context Large Language Models (LLMs) face significant memory bottlenecks during inference due to the linear growth of key-value ( KV) cache with sequence length.<n>While individual optimization techniques like KV cache quantization, chunked prefill, and model weight quantization have shown promise, their joint effects and optimal configurations for edge deployment remain underexplored.
• Score: 2.8485297992257017
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Long-context Large Language Models (LLMs) face significant memory bottlenecks during inference due to the linear growth of key-value (KV) cache with sequence length. While individual optimization techniques like KV cache quantization, chunked prefill, and model weight quantization have shown promise, their joint effects and optimal configurations for edge deployment remain underexplored. We introduce KV Pareto, a systems-level framework that systematically maps the trade-off frontier between total memory consumption and task accuracy across these three complementary optimization techniques. Our framework evaluates multiple LLM architectures (Qwen, Llama, Mistral) with varying KV quantization schemes (int2/4/8, mixed-precision), granularities (per-token, per-tensor, per-block), and 4-bit weight quantization via AWQ. Our framework identifies model-specific Pareto-optimal configurations that achieve 68-78% total memory reduction with minimal (1-3%) accuracy degradation on long-context tasks. We additionally verify the selected frontiers on additional benchmarks of Needle-in-a-Haystack, GSM8k and MMLU as well as extended context lengths of up to 128k to demonstrate the practical need of joint optimization for efficient LLM inference.

Related papers

• Fast KVzip: Efficient and Accurate LLM Inference with Gated KV Eviction [50.99402504483692]
  We propose a novel gating-based KV cache eviction method for frozen-weight language models.<n>Our approach integrates seamlessly into both the prefill and decoding stages.<n>Experiments show that our method maintains near-lossless performance while evicting up to 70% of the KV cache.
  arXiv  Detail & Related papers  (2026-01-25T03:07:54Z)
• Enhancing Large Multimodal Models with Adaptive Sparsity and KV Cache Compression [15.87033491172396]
  Large multimodal models (LMMs) have advanced significantly by integrating visual encoders with extensive language models, enabling robust reasoning capabilities.<n> compressing LMMs for deployment on edge devices remains a critical challenge.<n>We propose an adaptive search algorithm that optimize sparsity and KV cache compression to enhance LMM efficiency.
  arXiv  Detail & Related papers  (2025-07-28T08:27:40Z)
• KVmix: Gradient-Based Layer Importance-Aware Mixed-Precision Quantization for KV Cache [13.662270631753135]
  Quantization can effectively alleviate the memory pressure caused by KV Cache.<n>We propose a novel mixed-precision quantization method for KV Cache named KVmix.
  arXiv  Detail & Related papers  (2025-05-18T07:04:53Z)
• KVTuner: Sensitivity-Aware Layer-Wise Mixed-Precision KV Cache Quantization for Efficient and Nearly Lossless LLM Inference [40.97781175723418]
  KV cache quantization can improve Large Language Models inference throughput and latency in long contexts.<n>Current methods have three unsolved issues: overlooking layer-wise sensitivity to KV cache quantization, high overhead of online fine-grained decision-making, and low flexibility to different LLMs and constraints.<n>We propose a simple yet effective framework KVTuner to adaptively search for the optimal hardware-friendly layer-wise KV quantization precision pairs for coarse-grained KV cache.
  arXiv  Detail & Related papers  (2025-02-06T15:26:26Z)
• QuantSpec: Self-Speculative Decoding with Hierarchical Quantized KV Cache [67.84112700032007]
  Large Language Models (LLMs) are increasingly being deployed on edge devices for long-context settings.<n>In these scenarios, the Key-Value ( KV) cache is the primary bottleneck in terms of both GPU memory and latency.<n>We propose a novel self-speculative decoding framework, QuantSpec, where the draft model shares the architecture of the target model but employs a hierarchical 4-bit quantized KV cache and 4-bit quantized weights for acceleration.
  arXiv  Detail & Related papers  (2025-02-05T20:43:48Z)
• ThinK: Thinner Key Cache by Query-Driven Pruning [63.13363917871414]
  Large Language Models (LLMs) have revolutionized the field of natural language processing, achieving unprecedented performance across a variety of applications.<n>This paper focuses on the long-context scenario, addressing the inefficiencies in KV cache memory consumption during inference.<n>We propose ThinK, a novel query-dependent KV cache pruning method designed to minimize attention weight loss while selectively pruning the least significant channels.
  arXiv  Detail & Related papers  (2024-07-30T17:59:08Z)
• Model Tells You Where to Merge: Adaptive KV Cache Merging for LLMs on Long-Context Tasks [21.815661269986425]
  We propose a novel KV cache merging approach, called KVMerger, to achieve adaptive KV cache compression for long-context tasks. Our approach is inspired by the intriguing observation that key states exhibit high similarity at the token level within a single sequence. We conduct extensive experiments to demonstrate the effectiveness of KVMerger for long-context tasks under constrained memory budgets.
  arXiv  Detail & Related papers  (2024-07-11T12:50:42Z)
• LoCoCo: Dropping In Convolutions for Long Context Compression [77.26610232994508]
  This paper presents a novel approach, Dropping In Convolutions for Long Context Compression (LoCoCo) LoCoCo employs only a fixed-size Key-Value ( KV) cache, and can enhance efficiency in both inference and fine-tuning stages.
  arXiv  Detail & Related papers  (2024-06-08T01:35:11Z)
• SliM-LLM: Salience-Driven Mixed-Precision Quantization for Large Language Models [63.118592279833656]
  Post-training quantization (PTQ) is an effective technique for compressing large language models (LLMs)<n>We propose SliM-LLM, a salience-driven mixed-precision quantization framework that allocates bit-widths at the group-wise.<n> Experiments show that SliM-LLM achieves superior performance across various LLMs at low bit-widths.
  arXiv  Detail & Related papers  (2024-05-23T16:21:48Z)
• KVQuant: Towards 10 Million Context Length LLM Inference with KV Cache Quantization [67.74400574357472]
  LLMs are seeing growing use for applications which require large context windows, and with these large context windows KV cache activations surface as the dominant contributor to memory consumption during inference.<n> Quantization is a promising approach for compressing KV cache activations; however, existing solutions fail to represent activations accurately in sub-4-bit precision.<n>Our work, KVQuant, facilitates low precision KV cache quantization by incorporating several novel methods.
  arXiv  Detail & Related papers  (2024-01-31T18:58:14Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.