# VL-Cache: Sparsity and Modality-Aware KV Cache Compression for Vision-Language Model Inference Acceleration

arXiv: https://arxiv.org/abs/2410.23317 · HTML: https://mlsanity.com/paper/2410.23317/

## Key facts

- Cited by 43 papers in the map (data through 14 Oct 2026).
- Placed in Visual token pruning and efficiency, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 33 of its 40 references that are in the map.
- First posted on arXiv 29 Oct 2024.

## Placed in

- Multimodal LLMs / vision-language models / [Visual token pruning and efficiency](https://mlsanity.com/mllm/visual-token-pruning-and-efficiency/)
- Efficient LLM inference and serving / [Visual token pruning in multimodal models](https://mlsanity.com/inference/visual-token-pruning-in-multimodal-models/)

## Builds on (most cited first, 30 of 33)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Visual Instruction Tuning](https://mlsanity.com/paper/2304.08485/) (2023)
- [Efficient Memory Management for Large Language Model Serving with PagedAttention](https://mlsanity.com/paper/2309.06180/) (2023)
- [Improved Baselines with Visual Instruction Tuning](https://mlsanity.com/paper/2310.03744/) (2023)
- [Mistral 7B](https://mlsanity.com/paper/2310.06825/) (2023)
- [LLaVA-OneVision: Easy Visual Task Transfer](https://mlsanity.com/paper/2408.03326/) (2024)
- [FlashAttention-2: Faster Attention with Better Parallelism and Work Partitioning](https://mlsanity.com/paper/2307.08691/) (2023)
- [Qwen-VL: A Versatile Vision-Language Model for Understanding, Localization, Text Reading, and Beyond](https://mlsanity.com/paper/2308.12966/) (2023)
- [Efficient Streaming Language Models with Attention Sinks](https://mlsanity.com/paper/2309.17453/) (2023)
- [Towards VQA Models That Can Read](https://mlsanity.com/paper/1904.08920/) (2019)
- [MathVista: Evaluating Mathematical Reasoning of Foundation Models in Visual Contexts](https://mlsanity.com/paper/2310.02255/) (2023)
- [ChartQA: A Benchmark for Question Answering about Charts with Visual and Logical Reasoning](https://mlsanity.com/paper/2203.10244/) (2022)
- [Microsoft COCO Captions: Data Collection and Evaluation Server](https://mlsanity.com/paper/1504.00325/) (2015)
- [GQA: Training Generalized Multi-Query Transformer Models from Multi-Head Checkpoints](https://mlsanity.com/paper/2305.13245/) (2023)
- [DocVQA: A Dataset for VQA on Document Images](https://mlsanity.com/paper/2007.00398/) (2020)
- [A Survey on Evaluation of Large Language Models](https://mlsanity.com/paper/2307.03109/) (2023)
- [SnapKV: LLM Knows What You are Looking for Before Generation](https://mlsanity.com/paper/2404.14469/) (2024)
- [H2O: Heavy-Hitter Oracle for Efficient Generative Inference of Large Language Models](https://mlsanity.com/paper/2306.14048/) (2023)
- [An Image is Worth 1/2 Tokens After Layer 2: Plug-and-Play Inference Acceleration for Large Vision-Language Models](https://mlsanity.com/paper/2403.06764/) (2024)
- [KIVI: A Tuning-Free Asymmetric 2bit Quantization for KV Cache](https://mlsanity.com/paper/2402.02750/) (2024)
- [Quest: Query-Aware Sparsity for Efficient Long-Context LLM Inference](https://mlsanity.com/paper/2406.10774/) (2024)
- [Model Tells You What to Discard: Adaptive KV Cache Compression for LLMs](https://mlsanity.com/paper/2310.01801/) (2023)
- [Scissorhands: Exploiting the Persistence of Importance Hypothesis for LLM KV Cache Compression at Test Time](https://mlsanity.com/paper/2305.17118/) (2023)
- [PyramidKV: Dynamic KV Cache Compression based on Pyramidal Information Funneling](https://mlsanity.com/paper/2406.02069/) (2024)
- [InfiniGen: Efficient Generative Inference of Large Language Models with Dynamic KV Cache Management](https://mlsanity.com/paper/2406.19707/) (2024)
- [PyramidInfer: Pyramid KV Cache Compression for High-throughput LLM Inference](https://mlsanity.com/paper/2405.12532/) (2024)
- [PaLI-3 Vision Language Models: Smaller, Faster, Stronger](https://mlsanity.com/paper/2310.09199/) (2023)
- [Keyformer: KV Cache Reduction through Key Tokens Selection for Efficient Generative Inference](https://mlsanity.com/paper/2403.09054/) (2024)
- [No Token Left Behind: Reliable KV Cache Compression via Importance-Aware Mixed Precision Quantization](https://mlsanity.com/paper/2402.18096/) (2024)

## Built on by (most cited first, 30 of 43)

- [FrameFusion: Combining Similarity and Importance for Video Token Reduction on Large Vision Language Models](https://mlsanity.com/paper/2501.01986/) (2024)
- [AdaReTaKe: Adaptive Redundancy Reduction to Perceive Longer for Video-language Understanding](https://mlsanity.com/paper/2503.12559/) (2025)
- [Don't Just Chase "Highlighted Tokens" in MLLMs: Revisiting Visual Holistic Context Retention](https://mlsanity.com/paper/2510.02912/) (2025)
- [ReTaKe: Reducing Temporal and Knowledge Redundancy for Long Video Understanding](https://mlsanity.com/paper/2412.20504/) (2024)
- [MMInference: Accelerating Pre-filling for Long-Context VLMs via Modality-Aware Permutation Sparse Attention](https://mlsanity.com/paper/2504.16083/) (2025)
- [SpecVLM: Enhancing Speculative Decoding of Video LLMs via Verifier-Guided Token Pruning](https://mlsanity.com/paper/2508.16201/) (2025)
- [GUI-KV: Efficient GUI Agents via KV Cache with Spatio-Temporal Awareness](https://mlsanity.com/paper/2510.00536/) (2025)
- [VisionSelector: End-to-End Learnable Visual Token Compression for Efficient Multimodal LLMs](https://mlsanity.com/paper/2510.16598/) (2025)
- [Make Each Token Count: Towards Improving Long-Context Performance with KV Cache Eviction](https://mlsanity.com/paper/2605.09649/) (2026)
- [EVICPRESS: Joint KV-Cache Compression and Eviction for Efficient LLM Serving](https://mlsanity.com/paper/2512.14946/) (2025)
- [Efficient Inference for Large Vision-Language Models: Bottlenecks, Techniques, and Prospects](https://mlsanity.com/paper/2604.05546/) (2026)
- [Beyond Intermediate States: Explaining Visual Redundancy through Language](https://mlsanity.com/paper/2503.20540/) (2025)
- [METok: Multi-Stage Event-based Token Compression for Efficient Long Video Understanding](https://mlsanity.com/paper/2506.02850/) (2025)
- [HiViS: Hiding Visual Tokens from the Drafter for Speculative Decoding in Vision-Language Models](https://mlsanity.com/paper/2509.23928/) (2025)
- [V-Rex: Real-Time Streaming Video LLM Acceleration via Dynamic KV Cache Retrieval](https://mlsanity.com/paper/2512.12284/) (2025)
- [What to Keep, What to Forget: A Rate--Distortion View of Memory Compaction in LLMs and Agents](https://mlsanity.com/paper/2607.08032/) (2026)
- [CalibQuant: 1-Bit KV Cache Quantization for Multimodal LLMs](https://mlsanity.com/paper/2502.14882/) (2025)
- [AirCache: Activating Inter-modal Relevancy KV Cache Compression for Efficient Large Vision-Language Model Inference](https://mlsanity.com/paper/2503.23956/) (2025)
- [EffiVLM-BENCH: A Comprehensive Benchmark for Evaluating Training-Free Acceleration in Large Vision-Language Models](https://mlsanity.com/paper/2506.00479/) (2025)
- [Prefill-Time Intervention for Mitigating Hallucination in Large Vision-Language Models](https://mlsanity.com/paper/2604.25642/) (2026)
- [TriAxialKV: Toward Extreme Low-Precision KV-Cache Quantization for Agentic Inference Tasks](https://mlsanity.com/paper/2605.17170/) (2026)
- [Reflex: Real-Time VLA Control through Streaming Inference](https://mlsanity.com/paper/2607.14695/) (2026)
- [Blink: Dynamic Visual Token Resolution for Enhanced Multimodal Understanding](https://mlsanity.com/paper/2512.10548/) (2025)
- [ST-Lite: Training-Free KV Cache Compression with Spatio-Trajectory Guidance for Long-Horizon GUI Agents](https://mlsanity.com/paper/2603.00188/) (2026)
- [A Policy-Driven Runtime Layer for Agentic LLM Serving](https://mlsanity.com/paper/2605.27744/) (2026)
- [AURA: Action-Gated Memory for Robot Policies at Constant VRAM](https://mlsanity.com/paper/2606.02775/) (2026)
- [Visual Access Boundaries in Vision-Language Model Reasoning](https://mlsanity.com/paper/2607.12815/) (2026)
- [Learning Agent Execution for KV-Cache Management in Agentic Serving](https://mlsanity.com/paper/2608.14624/) (2026)
- [Slim attention: cut your context memory in half without loss -- K-cache is all you need for MHA](https://mlsanity.com/paper/2503.05840/) (2025)
- [STaR-KV: Spatio-Temporal Adaptive Re-weighting for KV Cache Compression in GUI Vision-Language Models](https://mlsanity.com/paper/2606.01790/) (2026)

## Abstract (arXiv)

Vision-Language Models (VLMs) have demonstrated impressive performance across a versatile set of tasks. A key challenge in accelerating VLMs is storing and accessing the large Key-Value (KV) cache that encodes long visual contexts, such as images or videos. While existing KV cache compression methods are effective for Large Language Models (LLMs), directly migrating them to VLMs yields suboptimal accuracy and speedup. To bridge the gap, we propose VL-Cache, a novel KV cache compression recipe tailored for accelerating VLM inference. In this paper, we first investigate the unique sparsity pattern of VLM attention by distinguishing visual and text tokens in prefill and decoding phases. Based on these observations, we introduce a layer-adaptive sparsity-aware cache budget allocation method that effectively distributes the limited cache budget across different layers, further reducing KV cache size without compromising accuracy. Additionally, we develop a modality-aware token scoring policy to better evaluate the token importance. Empirical results on multiple benchmark datasets demonstrate that retaining only 10% of KV cache achieves accuracy comparable to that with full cache. In a speed benchmark, our method accelerates end-to-end latency of generating 100 tokens by up to 2.33x and speeds up decoding by up to 7.08x, while reducing the memory footprint of KV cache in GPU by 90%.