TurboQuant: Online Vector Quantization with Near-optimal Distortion Rate · HackerPapers discussions

TurboQuant: Online Vector Quantization with Near-optimal Distortion Rate — mentioned by sroussey

Thu, 30 Apr 2026 00:54:03 GMT

Actually, they published a year ago. Recent was being on official Google blog. https://arxiv.org/abs/2504.19874 https://research.google/blog/turboquant-redefining-ai-effici...

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

TurboQuant: Online Vector Quantization with Near-optimal Distortion Rate — mentioned by thethirdone

Tue, 21 Apr 2026 03:31:07 GMT

> To be fair, TurboQuant is also a purely theoretical paper. I just wanted to get this out and share. TurboQuant is not a purely theoretical paper. Section 4 "Experiments" (page 15) [0] has a bunch of figure based on actual GPU computations. [0]: https://arxiv.org/abs/2504.19874

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

TurboQuant: Online Vector Quantization with Near-optimal Distortion Rate — mentioned by lhl

Sun, 19 Apr 2026 09:32:02 GMT

BTW, a number of corrections. The TurboQuant paper was submitted to Arxiv back in April 2025: https://arxiv.org/abs/2504.19874 Current "TurboQuant" implementations are about 3.8X-4.9X on compression (w/ the higher end taking some significant hits of GSM8K performance) and with about 80-100% baseline speed (no improvement, regression): https://github.com/vllm-project/vllm/pull/38479 For those not paying attention, it's probably worth sending this and ongoing discussion for vLLM https://github.com/vllm-project/vllm/issues/38171 and llama.cpp through your summarizer of choice - TurboQuant is…

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

TurboQuant: Online Vector Quantization with Near-optimal Distortion Rate — HN submission

Fri, 03 Apr 2026 22:31:31 GMT

Show HN: Standalone TurboQuant KV Cache Inference

Implements TurboQuant (ICLR 2026, arXiv:2504.19874) KV cache compression directly inside a Transformers inference script. All algorithms are self-contained. Minimal dependencies. - uses https://huggingface.co/g023/Qwen3-1.77B-g023 as the demonstration model (throw model files in Qwen3-BEST folder)

3 points; 4 discussion replies on HN. Paper and source history. A submission is not an endorsement.

TurboQuant: Online Vector Quantization with Near-optimal Distortion Rate — mentioned by yorwba

Sun, 29 Mar 2026 17:09:27 GMT

It especially doesn't make sense considering that TurboQuant has been public on arXiv for almost a year: https://arxiv.org/abs/2504.19874 So it predates the late-2025 RAM price surge! https://pcpartpicker.com/trends/price/memory/ I think that either investors were extremely skittish that the stocks might crash and jumped at the first sign of trouble (creating a self-fulfilling prophecy) or they were trading on non-public information and analysts who don't have access to said information are reading too much into the temporal coincidence of the Google Research blog highlighting this paper.

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

TurboQuant: Online Vector Quantization with Near-optimal Distortion Rate — mentioned by amai

Sun, 29 Mar 2026 15:28:56 GMT

The PolarQuant paper: https://arxiv.org/abs/2502.02617 The TurboQuant paper: https://arxiv.org/abs/2504.19874 These elaborate ideas are not adequately represented in the article.

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

TurboQuant: Online Vector Quantization with Near-optimal Distortion Rate — mentioned by justsomeguy1996

Sun, 29 Mar 2026 11:10:54 GMT

I built a Python implementation of Google's TurboQuant paper (ICLR 2026) for vector search. The key thing that makes this different from PQ and other quantization methods: it's fully data-oblivious. The codebook is derived from math (not trained on your data), so you can add vectors online without ever rebuilding the index. Each vector encodes independently in ~4ms at d=1536. The repo reproduces the benchmarks from Section 4.4 of the paper — recall@1@k on GloVe (d=200) and OpenAI embeddings (d=1536, d=3072). At 4-bit on d=1536, you get 0.967 recall@1@1 with 8x compression. At 2-bit, 0.862…

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

TurboQuant: Online Vector Quantization with Near-optimal Distortion Rate — mentioned by T-A

Sat, 28 Mar 2026 13:19:32 GMT

> Google announced Turboquant[1] today Most commentators seem to have missed that the TurboQuant paper was posted in April 2025: https://arxiv.org/abs/2504.19874 So it's been public knowledge for almost year, and internal Google knowledge for longer. Did Google reduce its capital expenditure plans over the past year? Let's ask Google: Google is expecting to see a capex of between $175-185 billion in 2026, approximately double that of 2025. https://www.datacenterdynamics.com/en/news/google-estimates-...

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

TurboQuant: Online Vector Quantization with Near-optimal Distortion Rate — mentioned by fc417fc802

Thu, 26 Mar 2026 14:33:14 GMT

It looks nice! Fair enough about QJL - it seems to be nothing more than an unbiasing measure anyway. I'm not sure if it's my own misunderstanding or if the paper [0] has something of an error. Section 3.1 starts out to the effect "let x be on the unit hypersphere" (but I'm fairly certain it's actually not). Neither algorithm 1 nor algorithm 2 show a normalization step prior to rotating x. Algorithm 2 line 8 shows that the scalar returned is actually the magnitude of the residual without accounting for QJL. Anyway I'm pretty sure the authors inadvertently omitted that detail which really had…

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

TurboQuant: Online Vector Quantization with Near-optimal Distortion Rate — mentioned by htrp

Wed, 25 Mar 2026 22:13:17 GMT

The actual paper from April 2025 TurboQuant: Online Vector Quantization with Near-optimal Distortion Rate https://arxiv.org/abs/2504.19874

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

TurboQuant: Online Vector Quantization with Near-optimal Distortion Rate — mentioned by wbsun

Wed, 25 Mar 2026 21:38:17 GMT

The blog is new but the paper was submitted almost one year ago: https://arxiv.org/abs/2504.19874. Anyone has ideas if this is already implemented in many models (at least Gemini, I guess)? If that's the case, can I expect cheaper RAM for my computer :D

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

TurboQuant: Online Vector Quantization with Near-optimal Distortion Rate — HN submission

Wed, 25 Mar 2026 15:55:59 GMT

Google's extreme AI compression paper was on arXiv since April 2025

2 points; 1 discussion reply on HN. Paper and source history. A submission is not an endorsement.

TurboQuant: Online Vector Quantization with Near-optimal Distortion Rate — mentioned by fc417fc802

Wed, 25 Mar 2026 12:49:16 GMT

Yeah that's odd. It seems like you'd want an n-1 dimensional grid on the surface of the unit sphere rather than an n dimensional grid within which the sphere resides. Looking at the paper (https://arxiv.org/abs/2504.19874) they cite earlier work that does exactly that. They object that grid projection and binary search perform exceptionally poorly on the GPU. I don't think they're using a regular grid as depicted on the linked page. Equation 4 from the paper is how they compute centroids for the MSE optimal quantizer. Why specify MSE optimal you ask? Yeah so it turns out there's actually two…

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