# Zipage: Maintain High Request Concurrency for LLM Reasoning through Compressed PagedAttention

arXiv: https://arxiv.org/abs/2603.08743 · HTML: https://mlsanity.com/paper/2603.08743/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Placed in Test-time compute and efficient thinking, one of 17 directions in LLM reasoning.
- Builds on 16 of its 27 references that are in the map.
- First posted on arXiv 1 Mar 2026.

## Placed in

- LLM reasoning / [Test-time compute and efficient thinking](https://mlsanity.com/lr/test-time-compute-and-efficient-thinking/)
- Efficient LLM inference and serving / [Efficient reasoning and agent inference](https://mlsanity.com/inference/efficient-reasoning-and-agent-inference/)

## Builds on (most cited first, 16 of 16)

- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [Evaluating Large Language Models Trained on Code](https://mlsanity.com/paper/2107.03374/) (2021)
- [DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](https://mlsanity.com/paper/2501.12948/) (2025)
- [LiveCodeBench: Holistic and Contamination Free Evaluation of Large Language Models for Code](https://mlsanity.com/paper/2403.07974/) (2024)
- [GQA: Training Generalized Multi-Query Transformer Models from Multi-Head Checkpoints](https://mlsanity.com/paper/2305.13245/) (2023)
- [Fast Transformer Decoding: One Write-Head is All You Need](https://mlsanity.com/paper/1911.02150/) (2019)
- [PyramidKV: Dynamic KV Cache Compression based on Pyramidal Information Funneling](https://mlsanity.com/paper/2406.02069/) (2024)
- [From System 1 to System 2: A Survey of Reasoning Large Language Models](https://mlsanity.com/paper/2502.17419/) (2025)
- [Ada-KV: Optimizing KV Cache Eviction by Adaptive Budget Allocation for Efficient LLM Inference](https://mlsanity.com/paper/2407.11550/) (2024)
- [SARATHI: Efficient LLM Inference by Piggybacking Decodes with Chunked Prefills](https://mlsanity.com/paper/2308.16369/) (2023)
- [A Survey of Frontiers in LLM Reasoning: Inference Scaling, Learning to Reason, and Agentic Systems](https://mlsanity.com/paper/2504.09037/) (2025)
- [Shared Global and Local Geometry of Language Model Embeddings](https://mlsanity.com/paper/2503.21073/) (2025)
- [KV-Compress: Paged KV-Cache Compression with Variable Compression Rates per Attention Head](https://mlsanity.com/paper/2410.00161/) (2024)
- [Core Context Aware Transformers for Long Context Language Modeling](https://mlsanity.com/paper/2412.12465/) (2024)
- [PagedEviction: Structured Block-wise KV Cache Pruning for Efficient Large Language Model Inference](https://mlsanity.com/paper/2509.04377/) (2025)
- [G-KV: Decoding-Time KV Cache Eviction with Global Attention](https://mlsanity.com/paper/2512.00504/) (2025)

## Built on by (most cited first, 4 of 4)

- [vToken: Token-Level Virtualization for Reclaimable KV Caches](https://mlsanity.com/paper/2608.13263/) (2026)
- [Keyless Attention: Value-Space Routing and Value-Only Caching for Efficient Transformers](https://mlsanity.com/paper/2606.21848/) (2026)
- [GrowPage: On-Demand KV Budgeting for Efficient LLM Reasoning Serving](https://mlsanity.com/paper/2609.03494/) (2026)
- [Does Execution Require Target KV Fidelity? A Mixed-Fidelity KV Runtime for LLM Serving](https://mlsanity.com/paper/2609.33536/) (2026)

## Abstract (arXiv)

With reasoning becoming the generative paradigm for large language models (LLMs), the memory bottleneck caused by KV cache during the decoding phase has become a critical factor limiting high-concurrency service. Although existing KV cache eviction methods address the memory issue, most of them are impractical for industrial-grade applications. This paper introduces Compressed PagedAttention, a method that combines token-wise KV cache eviction with PagedAttention. We propose a comprehensive scheduling strategy and support prefix caching and asynchronous compression for Compressed PagedAttention. Based on this, we have developed a high-concurrency LLM inference engine, Zipage. On large-scale mathematical reasoning tasks, Zipage achieves around 95% of the performance of Full KV inference engines while delivering over 2.1× speedup.