Revisiting RaBitQ and TurboQuant: A Symmetric Comparison of Methods, Theory, and Experiments

TL;DR

This paper compares RaBitQ and TurboQuant methods in methodology, theory, and experiments, revealing RaBitQ's superior performance and reproducibility issues in TurboQuant's reported results.

Contribution

It provides a unified comparison framework, clarifies differences, and highlights reproducibility problems in TurboQuant's experimental claims.

Findings

TurboQuant performs worse than RaBitQ in most tested settings.

Several TurboQuant reported results could not be reproduced from their implementation.

The paper clarifies shared structures and genuine differences between the methods.

Abstract

This technical note revisits the relationship between RaBitQ and TurboQuant under a unified comparison framework. We compare the two methods in terms of methodology, theoretical guarantees, and empirical performance, using a reproducible, transparent, and symmetric setup. Our results show that, despite the claimed advantage of TurboQuant, TurboQuant performs worse than RaBitQ in most tested settings of inner-product estimation, nearest-neighbor search and KV cache quantization. We further find that several reported runtime and recall results in the TurboQuant paper could not be reproduced from the released implementation under the stated configuration. Overall, this note clarifies the shared structure and genuine differences between the two lines of work, while documenting reproducibility issues in the experimental results reported by the TurboQuant paper.