KVQuant: Towards 10 Million Context Length LLM Inference with KV Cache Quantization
Reviewed by Pithpith:ZKGBG2ZCopen to challenge →
read the original abstract
LLMs are seeing growing use for applications which require large context windows, and with these large context windows KV cache activations surface as the dominant contributor to memory consumption during inference. Quantization is a promising approach for compressing KV cache activations; however, existing solutions fail to represent activations accurately in sub-4-bit precision. Our work, KVQuant, facilitates low precision KV cache quantization by incorporating several novel methods: (i) Per-Channel Key Quantization, where we adjust the dimension along which we quantize the Key activations to better match the distribution; (ii) Pre-RoPE Key Quantization, where we quantize Key activations before the rotary positional embedding to mitigate its impact on quantization; (iii) Non-Uniform KV Cache Quantization, where we derive per-layer sensitivity-weighted non-uniform datatypes that better represent the distributions; and (iv) Per-Vector Dense-and-Sparse Quantization, where we isolate outliers separately for each vector to minimize skews in quantization ranges. By applying our method to the LLaMA, Llama-2, Llama-3, and Mistral models, we achieve < 0.1 perplexity degradation with 3-bit quantization on both Wikitext-2 and C4, outperforming existing approaches. Our method enables serving LLaMA-7B with a context length of up to 1 million on a single A100-80GB GPU and up to 10 million on an 8-GPU system. We develop custom CUDA kernels for KVQuant, showing that we can achieve up to ~1.7x speedups, compared to baseline fp16 matrix-vector multiplications, for the LLaMA-7B model.
This paper has not been read by Pith yet.
Forward citations
Cited by 26 Pith papers
-
Alignment Collapse Under KV Cache Quantization: Diagnosis and Mitigation
KV cache quantization silently erodes LLM safety alignment via vulnerable low-dimensional subspaces, diagnosed by Per-Channel Reduction into three failure modes and mitigated training-free with up to 97% recovery.
-
HeadQ: Model-Visible Distortion and Score-Space Correction for KV-Cache Quantization
HeadQ removes 84-94% of excess perplexity from 2-bit key quantization by storing low-rank residuals in a calibration-learned query basis for score-space correction and using A²-weighted distortion for values.
-
STAR-KV: Low-Rank KV Cache Compression via Soft Thresholding for Adaptive Rank Control
STAR-KV applies differentiable soft thresholding for per-head and per-block adaptive low-rank KV cache compression, combined with hybrid decomposition and low-rank-aware quantization, achieving up to 75% compression a...
-
VORT: Adaptive Power-Law Memory for NLP Transformers
VORT assigns learnable fractional orders to tokens and approximates their power-law retention kernels via sum-of-exponentials for efficient long-range dependency modeling in transformers.
-
Sequential KV Cache Compression via Probabilistic Language Tries: Beyond the Per-Vector Shannon Limit
Sequential KV compression via probabilistic language tries and predictive delta coding achieves 3.3-4.3 bits per token entropy, yielding up to 914x better ratios than TurboQuant even with large overhead.
-
FlashAttention-3: Fast and Accurate Attention with Asynchrony and Low-precision
FlashAttention-3 achieves 1.5-2x speedup on H100 GPUs for attention, reaching 740 TFLOPs/s (75% utilization) in FP16 and near 1.2 PFLOPs/s in FP8 while cutting numerical error by 2.6x versus baseline FP8 attention.
-
DeepSeek-V2: A Strong, Economical, and Efficient Mixture-of-Experts Language Model
DeepSeek-V2 delivers top-tier open-source LLM performance using only 21B active parameters by compressing the KV cache 93.3% and cutting training costs 42.5% via MLA and DeepSeekMoE.
-
From Rigid to Dynamic: Entropy-Guided Adaptive Inference for Long-Context LLMs
EntropyInfer adaptively allocates inference compute using per-head attention entropy for rigid/dynamic classification during prefilling and compresses KV cache with generated tokens, achieving up to 2.39x speedup on l...
-
Do Value Vectors in Deep Layers Need Context from the Residual Stream?
Deeper transformer layers benefit from context-free token-specific value vectors in a Bank of Values lookup table, improving performance over standard attention with less compute.
-
Adaptive Mass-Segmented KV Compression for Long-Context Reasoning
AMS KV compression adaptively partitions the cache by attention mass regions and assigns quotas to protect contiguous reasoning blocks during long-context LLM inference.
-
Runtime-Certified Bounded-Error Quantized Attention
A tiered KV cache architecture computes per-head per-step error bounds on quantized attention and uses adaptive fallback to guarantee bounded or exact outputs relative to FP16 reference.
-
SPHERICAL KV: Angle-Domain Attention and Rate-Distortion Retention for Efficient Long-Context Inference
Spherical KV combines angle-domain attention using spherical key codes with rate-distortion retention to cut KV cache residency and HBM traffic while keeping a paged, fusion-friendly decode path.
-
SPHERICAL KV: Angle-Domain Attention and Rate-Distortion Retention for Efficient Long-Context Inference
Spherical KV introduces angle-domain attention with spherical key parameterization and rate-distortion retention to cut KV cache residency while preserving efficient paged decoding.
-
HeadQ: Model-Visible Distortion and Score-Space Correction for KV-Cache Quantization
HeadQ reduces 84-94% of excess perplexity in 2-bit key quantization by adding low-rank logit corrections in a calibration-learned query basis, with further gains from an A^2-weighted value policy.
-
WindowQuant: Mixed-Precision KV Cache Quantization based on Window-Level Similarity for VLMs Inference Optimization
WindowQuant performs window-adaptive mixed-precision KV cache quantization guided by similarity to the text prompt, with reordering to enable efficient inference in VLMs.
-
PolyKV: A Shared Asymmetrically-Compressed KV Cache Pool for Multi-Agent LLM Inference
A single shared asymmetrically compressed KV cache pool enables up to 15 concurrent LLM agents with 2.91x compression, 97.7% memory reduction, and only +0.57% perplexity increase on Llama-3-8B.
-
EpiCache: Episodic KV Cache Management for Long-Term Conversation on Resource-Constrained Environments
EpiCache clusters long conversation history into coherent episodes for per-episode KV cache eviction, delivering up to 30% accuracy gains and 3.7x peak memory reduction on LongConvQA tasks under fixed budgets.
-
TurboQuant: Online Vector Quantization with Near-optimal Distortion Rate
TurboQuant achieves near-optimal vector quantization distortion for both MSE and inner products via random rotation and per-coordinate scalar quantization, with a formal proof that it matches lower bounds within a fac...
-
KIVI: A Tuning-Free Asymmetric 2bit Quantization for KV Cache
KIVI applies asymmetric 2-bit quantization to KV cache with per-channel keys and per-token values, reducing memory 2.6x and boosting throughput up to 3.47x with near-identical quality on Llama, Falcon, and Mistral.
-
SGLang: Efficient Execution of Structured Language Model Programs
SGLang is a new system that speeds up structured LLM programs by up to 6.4x using RadixAttention for KV cache reuse and compressed finite state machines for output decoding.
-
ASVD: Activation-aware Singular Value Decomposition for Compressing Large Language Models
ASVD compresses LLMs by 10-30% and KV caches by 50% via activation-aware SVD that absorbs outliers into transformed weights and calibrates per-layer sensitivity.
-
GSRQ: Gain-Shape Residual Quantization for Sub-1-bit KV Cache
GSRQ applies a gain-shape variant of K-means inside residual quantization to improve directional fidelity, raising LongBench accuracy from 11.34 to 33.54 at 1-bit on LLaMA-3-8B.
-
HeadQ: Model-Visible Distortion and Score-Space Correction for KV-Cache Quantization
HeadQ applies score-space logit corrections for keys and attention-weighted surrogates for values to KV-cache quantization, removing 84-94% of excess perplexity in 2-bit key experiments across six models.
-
Influence-Inspired Spectral Rotations for Extreme Low-Bit LLM Quantization
A WHT rotation plus per-coordinate activation-energy rescaling before auto-round quantization lowers WikiText-2 perplexity 15-58% versus vanilla auto-round at W2A16 on models from 135M to 1.5B parameters.
-
Protection Is (Nearly) All You Need: Structural Protection Dominates Scoring in Globally Capped KV Eviction
Structural protection of boundary tokens in globally capped KV cache eviction recovers 69-90% of full-cache quality at 13% retention and dominates differences among scoring policies.
-
A Survey on Efficient Inference for Large Language Models
The paper surveys techniques to speed up and reduce the resource needs of LLM inference, organized by data-level, model-level, and system-level changes, with comparative experiments on representative methods.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.