A Super Fast K-means for Indexing Vector Embeddings

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Kuffo, Leonardo, Hepkema, Sven, Boncz, Peter
Format: Preprint
Publié: 2026
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866915877636538368
author Kuffo, Leonardo
Hepkema, Sven
Boncz, Peter
author_facet Kuffo, Leonardo
Hepkema, Sven
Boncz, Peter
contents We present SuperKMeans: a k-means variant designed for clustering collections of high-dimensional vector embeddings. SuperKMeans' clustering is up to 7x faster than FAISS and Scikit-Learn on modern CPUs and up to 4x faster than cuVS on GPUs (Figure 1), while maintaining the quality of the resulting centroids for vector similarity search tasks. SuperKMeans acceleration comes from reducing data-access and compute overhead by reliably and efficiently pruning dimensions that are not needed to assign a vector to a centroid. Furthermore, we present Early Termination by Recall, a novel mechanism that early-terminates k-means when the quality of the centroids for retrieval tasks stops improving across iterations. In practice, this further reduces runtimes without compromising retrieval quality. We open-source our implementation at https://github.com/cwida/SuperKMeans
format Preprint
id arxiv_https___arxiv_org_abs_2603_20009
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle A Super Fast K-means for Indexing Vector Embeddings
Kuffo, Leonardo
Hepkema, Sven
Boncz, Peter
Machine Learning
Databases
Information Retrieval
We present SuperKMeans: a k-means variant designed for clustering collections of high-dimensional vector embeddings. SuperKMeans' clustering is up to 7x faster than FAISS and Scikit-Learn on modern CPUs and up to 4x faster than cuVS on GPUs (Figure 1), while maintaining the quality of the resulting centroids for vector similarity search tasks. SuperKMeans acceleration comes from reducing data-access and compute overhead by reliably and efficiently pruning dimensions that are not needed to assign a vector to a centroid. Furthermore, we present Early Termination by Recall, a novel mechanism that early-terminates k-means when the quality of the centroids for retrieval tasks stops improving across iterations. In practice, this further reduces runtimes without compromising retrieval quality. We open-source our implementation at https://github.com/cwida/SuperKMeans
title A Super Fast K-means for Indexing Vector Embeddings
topic Machine Learning
Databases
Information Retrieval
url https://arxiv.org/abs/2603.20009