# Challenges in Deploying Long-Context Transformers: A Theoretical Peak Performance Analysis

arXiv: https://arxiv.org/abs/2405.08944 · HTML: https://mlsanity.com/paper/2405.08944/

## Key facts

- Cited by 44 papers in the map (data through 14 Oct 2026).
- Placed in Software engineering coding agents, one of 16 directions in LLM agents and tool use.
- Builds on 28 of its 31 references that are in the map.
- First posted on arXiv 14 May 2024.

## Placed in

- LLM agents and tool use / [Software engineering coding agents](https://mlsanity.com/agents/software-engineering-coding-agents/)
- Efficient LLM inference and serving / [Early exit and edge inference](https://mlsanity.com/inference/early-exit-and-edge-inference/)

## Builds on (most cited first, 28 of 28)

- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Scaling Laws for Neural Language Models](https://mlsanity.com/paper/2001.08361/) (2020)
- [Efficient Memory Management for Large Language Model Serving with PagedAttention](https://mlsanity.com/paper/2309.06180/) (2023)
- [Mamba: Linear-Time Sequence Modeling with Selective State Spaces](https://mlsanity.com/paper/2312.00752/) (2023)
- [Qwen Technical Report](https://mlsanity.com/paper/2309.16609/) (2023)
- [Mixtral of Experts](https://mlsanity.com/paper/2401.04088/) (2024)
- [Fast Inference from Transformers via Speculative Decoding](https://mlsanity.com/paper/2211.17192/) (2022)
- [DeepSeek-V2: A Strong, Economical, and Efficient Mixture-of-Experts Language Model](https://mlsanity.com/paper/2405.04434/) (2024)
- [Accelerating Large Language Model Decoding with Speculative Sampling](https://mlsanity.com/paper/2302.01318/) (2023)
- [SnapKV: LLM Knows What You are Looking for Before Generation](https://mlsanity.com/paper/2404.14469/) (2024)
- [Analyzing Multi-Head Self-Attention: Specialized Heads Do the Heavy Lifting, the Rest Can Be Pruned](https://mlsanity.com/paper/1905.09418/) (2019)
- [H2O: Heavy-Hitter Oracle for Efficient Generative Inference of Large Language Models](https://mlsanity.com/paper/2306.14048/) (2023)
- [Yi: Open Foundation Models by 01.AI](https://mlsanity.com/paper/2403.04652/) (2024)
- [KIVI: A Tuning-Free Asymmetric 2bit Quantization for KV Cache](https://mlsanity.com/paper/2402.02750/) (2024)
- [Model Tells You What to Discard: Adaptive KV Cache Compression for LLMs](https://mlsanity.com/paper/2310.01801/) (2023)
- [Confident Adaptive Language Modeling](https://mlsanity.com/paper/2207.07061/) (2022)
- [LayerSkip: Enabling Early Exit Inference and Self-Speculative Decoding](https://mlsanity.com/paper/2404.16710/) (2024)
- [LongT5: Efficient Text-To-Text Transformer for Long Sequences](https://mlsanity.com/paper/2112.07916/) (2021)
- [Retrieval Head Mechanistically Explains Long-Context Factuality](https://mlsanity.com/paper/2404.15574/) (2024)
- [Data Engineering for Scaling Language Models to 128K Context](https://mlsanity.com/paper/2402.10171/) (2024)
- [You Only Cache Once: Decoder-Decoder Architectures for Language Models](https://mlsanity.com/paper/2405.05254/) (2024)
- [Dynamic Memory Compression: Retrofitting LLMs for Accelerated Inference](https://mlsanity.com/paper/2403.09636/) (2024)
- [LongAlign: A Recipe for Long Context Alignment of Large Language Models](https://mlsanity.com/paper/2401.18058/) (2024)
- [TriForce: Lossless Acceleration of Long Sequence Generation with Hierarchical Speculative Decoding](https://mlsanity.com/paper/2404.11912/) (2024)
- [CoLT5: Faster Long-Range Transformers with Conditional Computation](https://mlsanity.com/paper/2303.09752/) (2023)
- [Infinite-LLM: Efficient LLM Service for Long Context with DistAttention and Distributed KVCache](https://mlsanity.com/paper/2401.02669/) (2024)
- [WKVQuant: Quantizing Weight and Key/Value Cache for Large Language Models Gains More](https://mlsanity.com/paper/2402.12065/) (2024)

## Built on by (most cited first, 30 of 44)

- [MInference 1.0: Accelerating Pre-filling for Long-Context LLMs via Dynamic Sparse Attention](https://mlsanity.com/paper/2407.02490/) (2024)
- [HippoRAG: Neurobiologically Inspired Long-Term Memory for Large Language Models](https://mlsanity.com/paper/2405.14831/) (2024)
- [FlexPrefill: A Context-Aware Sparse Attention Mechanism for Efficient Long-Sequence Inference](https://mlsanity.com/paper/2502.20766/) (2025)
- [A Simple and Effective L2 Norm-Based Strategy for KV Cache Compression](https://mlsanity.com/paper/2406.11430/) (2024)
- [Expected Attention: KV Cache Compression by Estimating Attention from Future Queries Distribution](https://mlsanity.com/paper/2510.00636/) (2025)
- [Palu: Compressing KV-Cache with Low-Rank Projection](https://mlsanity.com/paper/2407.21118/) (2024)
- [SCBench: A KV Cache-Centric Analysis of Long-Context Methods](https://mlsanity.com/paper/2412.10319/) (2024)
- [Block Transformer: Global-to-Local Language Modeling for Fast Inference](https://mlsanity.com/paper/2406.02657/) (2024)
- [The Sparse Frontier: Sparse Attention Trade-offs in Transformer LLMs](https://mlsanity.com/paper/2504.17768/) (2025)
- [Compute Or Load KV Cache? Why Not Both?](https://mlsanity.com/paper/2410.03065/) (2024)
- [SWAN-GPT: An Efficient and Scalable Approach for Long-Context Language Modeling](https://mlsanity.com/paper/2504.08719/) (2025)
- [xKV: Cross-Layer KV-Cache Compression via Aligned Singular Vector Extraction](https://mlsanity.com/paper/2503.18893/) (2025)
- [KVzap: Fast, Adaptive, and Faithful KV Cache Pruning](https://mlsanity.com/paper/2601.07891/) (2026)
- [FreeKV: Boosting KV Cache Retrieval for Efficient LLM Inference](https://mlsanity.com/paper/2505.13109/) (2025)
- [SlimInfer: Accelerating Long-Context LLM Inference via Dynamic Token Pruning](https://mlsanity.com/paper/2508.06447/) (2025)
- [PackCache: A Training-Free Acceleration Method for Unified Autoregressive Video Generation via Compact KV-Cache](https://mlsanity.com/paper/2601.04359/) (2026)
- [Interpolating Video-LLMs: Toward Longer-sequence LMMs in a Training-free Manner](https://mlsanity.com/paper/2409.12963/) (2024)
- [Analog In-Memory Computing Attention Mechanism for Fast and Energy-Efficient Large Language Models](https://mlsanity.com/paper/2409.19315/) (2024)
- [SideQuest: Model-Driven KV Cache Management for Long-Horizon Agentic Reasoning](https://mlsanity.com/paper/2602.22603/) (2026)
- [Let the Code LLM Edit Itself When You Edit the Code](https://mlsanity.com/paper/2407.03157/) (2024)
- [Re-ttention: Ultra Sparse Visual Generation via Attention Statistical Reshape](https://mlsanity.com/paper/2505.22918/) (2025)
- [KQ-SVD: Compressing the KV Cache with Provable Guarantees on Attention Fidelity](https://mlsanity.com/paper/2512.05916/) (2025)
- [freePruner: A Training-free Approach for Large Multimodal Model Acceleration](https://mlsanity.com/paper/2411.15446/) (2024)
- [Residual vector quantization for KV cache compression in large language model](https://mlsanity.com/paper/2410.15704/) (2024)
- [An LMM for Efficient Video Understanding via Reinforced Compression of Video Cubes](https://mlsanity.com/paper/2504.15270/) (2025)
- [CATransformers: Carbon Aware Transformers Through Joint Model-Hardware Optimization](https://mlsanity.com/paper/2505.01386/) (2025)
- [StreamAdapter: Efficient Test Time Adaptation from Contextual Streams](https://mlsanity.com/paper/2411.09289/) (2024)
- [HashEvict: A Pre-Attention KV Cache Eviction Strategy using Locality-Sensitive Hashing](https://mlsanity.com/paper/2412.16187/) (2024)
- [SparK: Query-Aware Unstructured Sparsity with Recoverable KV Cache Channel Pruning](https://mlsanity.com/paper/2508.15212/) (2025)
- [Q-Filters: Leveraging QK Geometry for Efficient KV Cache Compression](https://mlsanity.com/paper/2503.02812/) (2025)

## Abstract (arXiv)

Transformer-based long context generative models power emerging AI applications like hour-long video understanding and project-level coding agent. Deploying long context transformers (e.g., 100K to 10M tokens) is prohibitively expensive compared to short context (e.g., 4K tokens) model variants. Reducing the cost of long-context transformers is becoming a pressing research and engineering challenge starting from the year of 2024. This work describes a concurrent programming framework for quantitatively analyzing the efficiency challenges in serving multiple long-context requests under limited size of GPU high-bandwidth memory (HBM) regime. We give a detailed analysis of how all additional computational costs, compared to 4K context, trace back to textitone single source: the large size of the KV cache. We use a 34B GPT-3.5 level model of 50K context on A100 NVLink as a running example, and describe how its large KV cache causes four types of deployment challenges: (1) prefilling long inputs takes much longer compute time and GPU memory than short inputs; (2) after prefilling, the large KV cache residing on the GPU HBM substantially restricts the number of concurrent users being served; (3) during decoding, repeatedly reading the KV cache from HBM to SM largely increases latency; (4) when KV cache memory overflows, swapping it from HBM to DDR causes significant context switching latency. We use this framework to analyze existing works and identify possibilities of combining them to build end-to-end systems. Overall, this work offers a foundational framework for analyzing long context transformer deployment and identifies directions towards reducing the inference cost of 1M context to be as cheap as 4K.