SnapKV: LLM Knows What You are Looking for Before Generation

By Yuhong Li, Yingbing Huang, Bowen Yang, Bharat Venkitesh, Acyr F. Locatelli, Hanchen Ye, Tianle Cai, Patrick Lewis, Deming Chen. Neural Information Processing Systems, 2024

759 citations recorded by the source corpus.

Large Language Models (LLMs) have made remarkable progress in processing extensive contexts, with the Key-Value (KV) cache playing a vital role in enhancing their performance. However, the growth of the KV cache in response to increasing input length poses challenges to memory and time efficiency. To address this problem, this paper introduces SnapKV, an innovative and fine-tuning-free approach that efficiently minimizes KV cache size while still delivering comparable performance in real-world applications. We discover that each attention head in the model consistently focuses on specific prompt attention features during generation. Meanwhile, this robust pattern can be obtained from an 'observation' window located at the end of the prompts. Drawing on this insight, SnapKV automatically compresses KV caches by selecting clustered important KV positions for each attention head. Our approach significantly reduces the growing computational overhead and memory footprint when processing long input sequences. Specifically, SnapKV achieves a consistent decoding speed with a 3.6x increase in generation speed and an 8.2x enhancement in memory efficiency compared to the baseline when processing inputs of 16K tokens. At the same time, it maintains comparable performance to the baseline models across 16 long sequence datasets. Moreover, SnapKV can process up to 380K context tokens on a single A100-80GB GPU using HuggingFace implementation with minor changes, exhibiting only a negligible accuracy drop in the Needle-in-a-Haystack test. Further comprehensive studies suggest SnapKV's potential for practical applications.

View Paper: https://www.semanticscholar.org/paper/1784c987e681d60c634765fe64c8d9c26f73d5ff

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Zefan Cai
Yichi Zhang
Tianyu Liu
Bofei Gao
Yuliang Liu
Baobao Chang
Wayne Xiong
Keming Lu

Citations in the corpus

PyramidKV: Dynamic KV Cache Compression based on Pyramidal Information Funneling (2024, methodology): “Previous work on KV cache compression [22, 34, 32] often allocates a fixed KV cache size across LLM layers.”
MInference 1.0: Accelerating Pre-filling for Long-Context LLMs via Dynamic Sparse Attention (2024, unknown): “Our experiments use four state-of-the-art long-context LLMs: LLaMA-3-8B-Instruct-262k 1 , LLaMA-3-8B-Instruct-1048k”
Mindscape-Aware Retrieval Augmented Generation for Improved Long Context Understanding (2025, unknown): “To capture the overarching semantics of long documents, recent work moves beyond token retention or KV-cache compression (Yang et al., 2024; Li et al., 2024; Xiao et al., 2023) toward abstractive compression and global memory (Qian et al., 2025; Behrouz et al., 2024).”