# PM-KVQ: Progressive Mixed-precision KV Cache Quantization for Long-CoT LLMs

arXiv: https://arxiv.org/abs/2505.18610 · HTML: https://mlsanity.com/paper/2505.18610/

## Key facts

- Cited by 22 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 24 of its 32 references that are in the map.
- First posted on arXiv 24 May 2025.

## 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 / [KV cache quantization](https://mlsanity.com/inference/kv-cache-quantization/)

## Builds on (most cited first, 24 of 24)

- [DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](https://mlsanity.com/paper/2501.12948/) (2025)
- [RoFormer: Enhanced Transformer with Rotary Position Embedding](https://mlsanity.com/paper/2104.09864/) (2021)
- [Efficient Streaming Language Models with Attention Sinks](https://mlsanity.com/paper/2309.17453/) (2023)
- [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)
- [AWQ: Activation-aware Weight Quantization for LLM Compression and Acceleration](https://mlsanity.com/paper/2306.00978/) (2023)
- [Instruction-Following Evaluation for Large Language Models](https://mlsanity.com/paper/2311.07911/) (2023)
- [DeepSeek-V2: A Strong, Economical, and Efficient Mixture-of-Experts Language Model](https://mlsanity.com/paper/2405.04434/) (2024)
- [H2O: Heavy-Hitter Oracle for Efficient Generative Inference of Large Language Models](https://mlsanity.com/paper/2306.14048/) (2023)
- [Extending Context Window of Large Language Models via Positional Interpolation](https://mlsanity.com/paper/2306.15595/) (2023)
- [Fast Transformer Decoding: One Write-Head is All You Need](https://mlsanity.com/paper/1911.02150/) (2019)
- [CVXPY: A Python-Embedded Modeling Language for Convex Optimization](https://mlsanity.com/paper/1603.00943/) (2016)
- [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)
- [QServe: W4A8KV4 Quantization and System Co-design for Efficient LLM Serving](https://mlsanity.com/paper/2405.04532/) (2024)
- [RedPajama: an Open Dataset for Training Large Language Models](https://mlsanity.com/paper/2411.12372/) (2024)
- [No Token Left Behind: Reliable KV Cache Compression via Importance-Aware Mixed Precision Quantization](https://mlsanity.com/paper/2402.18096/) (2024)
- [WKVQuant: Quantizing Weight and Key/Value Cache for Large Language Models Gains More](https://mlsanity.com/paper/2402.12065/) (2024)
- [IntactKV: Improving Large Language Model Quantization by Keeping Pivot Tokens Intact](https://mlsanity.com/paper/2403.01241/) (2024)
- [RotateKV: Accurate and Robust 2-Bit KV Cache Quantization for LLMs via Outlier-Aware Adaptive Rotations](https://mlsanity.com/paper/2501.16383/) (2025)
- [SKVQ: Sliding-window Key and Value Cache Quantization for Large Language Models](https://mlsanity.com/paper/2405.06219/) (2024)
- [Mixture of Attention Spans: Optimizing LLM Inference Efficiency with Heterogeneous Sliding-Window Lengths](https://mlsanity.com/paper/2406.14909/) (2024)
- [MixDQ: Memory-Efficient Few-Step Text-to-Image Diffusion Models with Metric-Decoupled Mixed Precision Quantization](https://mlsanity.com/paper/2405.17873/) (2024)
- [KVTuner: Sensitivity-Aware Layer-Wise Mixed-Precision KV Cache Quantization for Efficient and Nearly Lossless LLM Inference](https://mlsanity.com/paper/2502.04420/) (2025)

## Built on by (most cited first, 22 of 22)

- [R2R: Efficiently Navigating Divergent Reasoning Paths with Small-Large Model Token Routing](https://mlsanity.com/paper/2505.21600/) (2025)
- [ThinKV: Thought-Adaptive KV Cache Compression for Efficient Reasoning Models](https://mlsanity.com/paper/2510.01290/) (2025)
- [PAROAttention: Pattern-Aware ReOrdering for Efficient Sparse and Quantized Attention in Visual Generation Models](https://mlsanity.com/paper/2506.16054/) (2025)
- [OSCAR: Offline Spectral Covariance-Aware Rotation for 2-bit KV Cache Quantization](https://mlsanity.com/paper/2605.17757/) (2026)
- [MixKVQ: Query-Aware Mixed-Precision KV Cache Quantization for Long-Context Reasoning](https://mlsanity.com/paper/2512.19206/) (2025)
- [KVServe: Service-Aware KV Cache Compression for Communication-Efficient Disaggregated LLM Serving](https://mlsanity.com/paper/2605.13734/) (2026)
- [Mixed-Precision Quantization for Language Models: Techniques and Prospects](https://mlsanity.com/paper/2510.16805/) (2025)
- [SplitZip: Ultra Fast Lossless KV Compression for Disaggregated LLM Serving](https://mlsanity.com/paper/2605.01708/) (2026)
- [RateQuant: Optimal Mixed-Precision KV Cache Quantization via Rate-Distortion Theory](https://mlsanity.com/paper/2605.06675/) (2026)
- [TriAxialKV: Toward Extreme Low-Precision KV-Cache Quantization for Agentic Inference Tasks](https://mlsanity.com/paper/2605.17170/) (2026)
- [CacheFlow: Efficient LLM Serving with 3D-Parallel KV Cache Restoration](https://mlsanity.com/paper/2604.25080/) (2026)
- [RoPE-Aware Bit Allocation for KV-Cache Quantization](https://mlsanity.com/paper/2606.24033/) (2026)
- [Monte Carlo Estimation for KV Cache Eviction](https://mlsanity.com/paper/2610.07643/) (2026)
- [RotateAttention: RoPE-Aware Rotation and Range Rectification for INT4 Quantized Attention in Video Generation](https://mlsanity.com/paper/2607.02584/) (2026)
- [High-accuracy Low-Bit KV-Cache Quantization via Local Distribution Restoration](https://mlsanity.com/paper/2607.16248/) (2026)
- [Does Execution Require Target KV Fidelity? A Mixed-Fidelity KV Runtime for LLM Serving](https://mlsanity.com/paper/2609.33536/) (2026)
- [AvoKV-E: Payload-Aware KV Cache Eviction for Long Reasoning](https://mlsanity.com/paper/2610.03007/) (2026)
- [Spend Bytes on Breadth: Precision-Count Trade-offs for Decode-Time KV Compression in Long Chain-of-Thought Reasoning](https://mlsanity.com/paper/2610.05685/) (2026)
- [Read What Matters: Query-Adaptive Quantization for KV Caches](https://mlsanity.com/paper/2610.11245/) (2026)
- [KV4-RL: Unlocking Faster Asynchronous Rollouts for RLVR with INT4 KV Cache Quantization](https://mlsanity.com/paper/2610.14448/) (2026)
- [MultiAlloc: Budget-Aware Precision Allocation for KV Cache Quantization](https://mlsanity.com/paper/2610.18642/) (2026)
- [PN-Mix: Pair-Norm Guided Mixed-Precision Key Cache Quantization for Efficient Long-Output LLM Inference](https://mlsanity.com/paper/2610.19605/) (2026)

## Abstract (arXiv)

Recently, significant progress has been made in developing reasoning-capable Large Language Models (LLMs) through long Chain-of-Thought (CoT) techniques. However, this long-CoT reasoning process imposes substantial memory overhead due to the large Key-Value (KV) Cache memory overhead. Post-training KV Cache quantization has emerged as a promising compression technique and has been extensively studied in short-context scenarios. However, directly applying existing methods to long-CoT LLMs causes significant performance degradation due to the following two reasons: (1) Large cumulative error: Existing methods fail to adequately leverage available memory, and they directly quantize the KV Cache during each decoding step, leading to large cumulative quantization error. (2) Short-context calibration: Due to Rotary Positional Embedding (RoPE), the use of short-context data during calibration fails to account for the distribution of less frequent channels in the Key Cache, resulting in performance loss. We propose Progressive Mixed-Precision KV Cache Quantization (PM-KVQ) for long-CoT LLMs to address the above issues in two folds: (1) To reduce cumulative error, we design a progressive quantization strategy to gradually lower the bit-width of KV Cache in each block. Then, we propose block-wise memory allocation to assign a higher bit-width to more sensitive transformer blocks. (2) To increase the calibration length without additional overhead, we propose a new calibration strategy with positional interpolation that leverages short calibration data with positional interpolation to approximate the data distribution of long-context data. Extensive experiments on 7B-70B long-CoT LLMs show that PM-KVQ improves reasoning benchmark performance by up to 8% over SOTA baselines under the same memory budget and achieves 2.73-5.18x throughput over the original 16-bit LLMs.