A Note on TurboQuant and the Earlier DRIVE/EDEN Line of Work · HackerPapers discussions

A Note on TurboQuant and the Earlier DRIVE/EDEN Line of Work — HN submission

Sat, 02 May 2026 18:53:06 GMT

A Note on TurboQuant and the Earlier Eden Work

2 points; 0 discussion replies on HN. Paper and source history. A submission is not an endorsement.

A Note on TurboQuant and the Earlier DRIVE/EDEN Line of Work — mentioned by amitport

Sat, 02 May 2026 13:39:46 GMT

Hi, I’m the author of the post above. tl;dr TurboQuant is a recent paper from Google and NYU that has gained massive traction in mainstream media and the AI community. As implementations of TurboQuant are integrated into various popular projects, it is important to note its relation to EDEN quantization. TurboQuant is essentially a partial implementation of EDEN quantization (first work published in NeurIPS 2021, extention published on ICML 2022). The few differences that do exist make EDEN significantly better. We have also published a detailed comparative report here…

Paper and discussion history. A mention is not an endorsement.

A Note on TurboQuant and the Earlier DRIVE/EDEN Line of Work — mentioned by KnuthIsGod

Mon, 27 Apr 2026 04:56:53 GMT

https://arxiv.org/abs/2604.18555 "This note clarifies the relationship between the recent TurboQuant work and the earlier DRIVE (NeurIPS 2021) and EDEN (ICML 2022) schemes. DRIVE is a 1-bit quantizer that EDEN extended to any bits per coordinate; we refer to them collectively as EDEN. First, TurboQuant is a special case of EDEN obtained by fixing EDEN's scalar scale parameter to . EDEN supports both biased and unbiased quantization, each optimized by a different (chosen via methods described in the EDEN works). The fixed choice used by TurboQuant is generally suboptimal, although the optimal…

Paper and discussion history. A mention is not an endorsement.

A Note on TurboQuant and the Earlier DRIVE/EDEN Line of Work — mentioned by amitport

Mon, 27 Apr 2026 04:01:24 GMT

TurboQuant is a restricted version of EDEN quantization (NeurIPS 21, ICML 22). It lacks the optimal scale derivations, which makes the TurboQuant variant considerably less accurate than those works. We show this thoroughly in a new note at https://arxiv.org/abs/2604.18555. We were the first to introduce post-rotation distribution-aware quantization in 2021. This was later implemented in many fields, including federated learning, vector retrieval, databases, inference engines, and KV-cache. It would be appropriate to receive credit for this. Furthermore, it is baffling to see the name…

Paper and discussion history. A mention is not an endorsement.