# vTensor: Flexible Virtual Tensor Management for Efficient LLM Serving

arXiv: https://arxiv.org/abs/2407.15309 · HTML: https://mlsanity.com/paper/2407.15309/

## Key facts

- Cited by 15 papers in the map (data through 14 Oct 2026).
- Placed in KV cache optimization, one of 12 directions in Efficient LLM inference and serving.
- Builds on 29 of its 45 references that are in the map.
- First posted on arXiv 22 Jul 2024.

## Placed in

- Efficient LLM inference and serving / [KV cache optimization](https://mlsanity.com/inference/kv-cache-optimization/)

## Builds on (most cited first, 29 of 29)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [Efficient Memory Management for Large Language Model Serving with PagedAttention](https://mlsanity.com/paper/2309.06180/) (2023)
- [FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness](https://mlsanity.com/paper/2205.14135/) (2022)
- [FlashAttention-2: Faster Attention with Better Parallelism and Work Partitioning](https://mlsanity.com/paper/2307.08691/) (2023)
- [Deep Compression: Compressing Deep Neural Networks with Pruning, Trained Quantization and Huffman Coding](https://mlsanity.com/paper/1510.00149/) (2015)
- [SmoothQuant: Accurate and Efficient Post-Training Quantization for Large Language Models](https://mlsanity.com/paper/2211.10438/) (2022)
- [GQA: Training Generalized Multi-Query Transformer Models from Multi-Head Checkpoints](https://mlsanity.com/paper/2305.13245/) (2023)
- [Yi: Open Foundation Models by 01.AI](https://mlsanity.com/paper/2403.04652/) (2024)
- [Fast Transformer Decoding: One Write-Head is All You Need](https://mlsanity.com/paper/1911.02150/) (2019)
- [Splitwise: Efficient generative LLM inference using phase splitting](https://mlsanity.com/paper/2311.18677/) (2023)
- [Taming Throughput-Latency Tradeoff in LLM Inference with Sarathi-Serve](https://mlsanity.com/paper/2403.02310/) (2024)
- [Recent Advances in Natural Language Processing via Large Pre-Trained Language Models: A Survey](https://mlsanity.com/paper/2111.01243/) (2021)
- [Atom: Low-bit Quantization for Efficient and Accurate LLM Serving](https://mlsanity.com/paper/2310.19102/) (2023)
- [Prompt Cache: Modular Attention Reuse for Low-Latency Inference](https://mlsanity.com/paper/2311.04934/) (2023)
- [Cost-Efficient Large Language Model Serving for Multi-turn Conversations with CachedAttention](https://mlsanity.com/paper/2403.19708/) (2024)
- [M6-Rec: Generative Pretrained Language Models are Open-Ended Recommender Systems](https://mlsanity.com/paper/2205.08084/) (2022)
- [Inference without Interference: Disaggregate LLM Inference for Mixed Downstream Workloads](https://mlsanity.com/paper/2401.11181/) (2024)
- [vAttention: Dynamic Memory Management for Serving LLMs without PagedAttention](https://mlsanity.com/paper/2405.04437/) (2024)
- [OliVe: Accelerating Large Language Models via Hardware-friendly Outlier-Victim Pair Quantization](https://mlsanity.com/paper/2304.07493/) (2023)
- [LV-Eval: A Balanced Long-Context Benchmark with 5 Length Levels Up to 256K](https://mlsanity.com/paper/2402.05136/) (2024)
- [Infinite-LLM: Efficient LLM Service for Long Context with DistAttention and Distributed KVCache](https://mlsanity.com/paper/2401.02669/) (2024)
- [SQuant: On-the-Fly Data-Free Quantization via Diagonal Hessian Approximation](https://mlsanity.com/paper/2202.07471/) (2022)
- [ANT: Exploiting Adaptive Numerical Data Type for Low-bit Deep Neural Network Quantization](https://mlsanity.com/paper/2208.14286/) (2022)
- [Accelerating Sparse DNN Models without Hardware-Support via Tile-Wise Sparsity](https://mlsanity.com/paper/2008.13006/) (2020)
- [Dual-side Sparse Tensor Core](https://mlsanity.com/paper/2105.09564/) (2021)
- [GMLake: Efficient and Transparent GPU Memory Defragmentation for Large-scale DNN Training with Virtual Memory Stitching](https://mlsanity.com/paper/2401.08156/) (2024)
- [Accelerating Sparse DNNs Based on Tiled GEMM](https://mlsanity.com/paper/2402.10876/) (2024)
- [Evaluation of computational and energy performance in matrix multiplication algorithms on CPU and GPU using MKL, cuBLAS and SYCL](https://mlsanity.com/paper/2405.17322/) (2024)

## Built on by (most cited first, 15 of 15)

- [A Survey on Large Language Model Acceleration based on KV Cache Management](https://mlsanity.com/paper/2412.19442/) (2024)
- [Prism: Cost-Efficient Multi-LLM Serving via GPU Memory Ballooning](https://mlsanity.com/paper/2505.04021/) (2025)
- [LayerKV: Optimizing Large Language Model Serving with Layer-wise KV Cache Management](https://mlsanity.com/paper/2410.00428/) (2024)
- [M-ANT: Efficient Low-bit Group Quantization for LLMs via Mathematically Adaptive Numerical Type](https://mlsanity.com/paper/2502.18755/) (2025)
- [HeadInfer: Memory-Efficient LLM Inference by Head-wise Offloading](https://mlsanity.com/paper/2502.12574/) (2025)
- [COMET: Towards Partical W4A4KV4 LLMs Serving](https://mlsanity.com/paper/2410.12168/) (2024)
- [Towards Efficient Large Language Model Serving: A Survey on System-Aware KV Cache Optimization](https://mlsanity.com/paper/2607.08057/) (2026)
- [FastCache: Optimizing Multimodal LLM Serving through Lightweight KV-Cache Compression Framework](https://mlsanity.com/paper/2503.08461/) (2025)
- [HCAttention: Extreme KV Cache Compression via Heterogeneous Attention Computing for LLMs](https://mlsanity.com/paper/2507.19823/) (2025)
- [xLLM Technical Report](https://mlsanity.com/paper/2510.14686/) (2025)
- [Fuel Gauge: Estimating Chain-of-Thought Length Ahead of Time in Large Multimodal Models](https://mlsanity.com/paper/2603.10335/) (2026)
- [Comparative Characterization of KV Cache Management Strategies for LLM Inference](https://mlsanity.com/paper/2604.05012/) (2026)
- [VTC: DNN Compilation with Virtual Tensors for Data Movement Elimination](https://mlsanity.com/paper/2604.09558/) (2026)
- [Asymmetric Virtual Memory Paging for Hybrid Mamba-Transformer Inference](https://mlsanity.com/paper/2605.22416/) (2026)
- [FastEEP: Fast Intra-Instance Elastic Scaling for Expert-Parallel MoE Serving](https://mlsanity.com/paper/2610.18451/) (2026)

## Abstract (arXiv)

Large Language Models (LLMs) are widely used across various domains, processing millions of daily requests. This surge in demand poses significant challenges in optimizing throughput and latency while keeping costs manageable. The Key-Value (KV) cache, a standard method for retaining previous computations, makes LLM inference highly bounded by memory. While batching strategies can enhance performance, they frequently lead to significant memory fragmentation. Even though cutting-edge systems like vLLM mitigate KV cache fragmentation using paged Attention mechanisms, they still suffer from inefficient memory and computational operations due to the tightly coupled page management and computation kernels. This study introduces the vTensor, an innovative tensor structure for LLM inference based on GPU virtual memory management (VMM). vTensor addresses existing limitations by decoupling computation from memory defragmentation and offering dynamic extensibility. Our framework employs a CPU-GPU heterogeneous approach, ensuring efficient, fragmentation-free memory management while accommodating various computation kernels across different LLM architectures. Experimental results indicate that vTensor achieves an average speedup of 1.86x across different models, with up to 2.42x in multi-turn chat scenarios. Additionally, vTensor provides average speedups of 2.12x and 3.15x in kernel evaluation, reaching up to 3.92x and 3.27x compared to SGLang Triton prefix-prefilling kernels and vLLM paged Attention kernel, respectively. Furthermore, it frees approximately 71.25% (57GB) of memory on the NVIDIA A100 GPU compared to vLLM, enabling more memory-intensive workloads.