VQ-bench: A Composable Vector Quantization Framework

Ashwin Padaki; Amir Ingber; Edo Liberty

cs.AIJul 31, 2026

VQ-bench: A Composable Vector Quantization Framework

Authors: Ashwin Padaki, Amir Ingber, Edo Liberty

Organizations: University of Pennsylvania and Pinecone · Pinecone

Abstract

Vector quantization is an old problem but has recently become central to AI infrastructure. It is therefore experiencing a surge of renewed engineering and research activity. This paper provides a unified framework for developing and benchmarking new quantization algorithms. We describe 7 common conceptual quantization primitives and show how to compose them arbitrarily. We then re-express 25 common quantizers as pipelines of these primitives. Finally, we publish VQ-bench as open-source to be extended further and make reproducible benchmarks publicly available.

Explore similar work

RQ-MoE: Residual Quantization via Mixture of Experts for Efficient Input-Dependent Vector Compression
May 14, 2026Zhengjia Zhong, Shuyan Ke, Zaizhou Lin +3Calibration-Free Low-Bit Vector Quantization Codebooks
Block-Sphere Vector Quantization
May 19, 2026Heesang Ann, Joongkyu Lee, Min-hwan OhCalibration-Free Low-Bit Vector Quantization Extreme Low-Bit Quantization
Efficient VQ-QAT and Mixed Vector/Linear quantized Neural Networks
Apr 25, 2026Terry Gou, Puneet GuptaCalibration-Free Low-Bit Vector Quantization Context Compression

May 14, 2026 · cs.LGJ/K move · Enter open · S save

RQ-MoE: Residual Quantization via Mixture of Experts for Efficient Input-Dependent Vector Compression

Zhengjia Zhong, Shuyan Ke, Zaizhou Lin, Jiaqi Song+2

Vector quantization is a fundamental tool for compressing high-dimensional embeddings, yet existing multi-codebook methods rely on static codebooks that limit expressiveness under heterogeneous data geometry. While recent dynamic quantizers like QINCo adapt codebooks to individual inputs and improve expressiveness, their strict sequential dependencies create decoding bottlenecks. We propose Residual Quantization via Mixture of Experts (RQ-MoE), a framework combining a two-level MoE with dual-stream quantization to enable input-dependent codebook adaptation for efficient vector quantization. RQ-MoE enables dynamic codebook construction and decouples instruction from quantization, facilitating parallel decoding. Theoretically, we show that standard Residual Quantization and QINCo can be recovered as constrained special cases of RQ-MoE, and derive a guideline for setting expert dimensionality in RQ-MoE. Extensive experiments show that RQ-MoE achieves state-of-the-art or on-par performance in reconstruction and retrieval, while providing 6x-14x faster decoding than prior vector quantization methods. The implementation is available at https://github.com/KDEGroup/RQ-MoE.

Calibration-Free Low-Bit Vector Quantization Codebooks Quantization Experts Intrinsic Dimensionality Parallel

Open paper Ask about it PDF

Abstract

Explore similar work

RQ-MoE: Residual Quantization via Mixture of Experts for Efficient Input-Dependent Vector Compression

Block-Sphere Vector Quantization

Efficient VQ-QAT and Mixed Vector/Linear quantized Neural Networks