arXiv 2504.19874 TurboQuant: Online Vector Quantization with Near-optimal Distortion Rate

Amir Zandieh, Majid Daliri, Majid Hadian, Vahab Mirrokni. arXiv:2504.19874v1, submitted 2025-04-28, 25 pages, cs.LG. Abstract states the quality result used here verbatim: for KV cache quantization, absolute quality neutrality at 3.5 bits per channel and marginal quality degradation at 2.5 bits per channel. Also claims near-optimal distortion (within about 2.7x of the information-theoretic lower bound) and that the method is data-oblivious, so it needs no calibration. The HTML/PDF body did not load from this host on 2026-10-18; the abstract did. Author-reported research result from Google Research, so numbers taken from it are vendor_claim even though the venue is a paper. Accepted at ICLR 2026 per the Google Research blog of 2026-03-24.