Hackaday

TurboQuant: Reducing LLM Memory Usage With Vector Quantization

Large language models (LLMs) aren’t actually giant computer brains. Instead, they are effectively massive vector spaces in ...
GitHub

Near-optimal vector quantization for LLM KV cache compression.

Random rotation: Multiply the input vector by a fixed random orthogonal matrix. This makes each coordinate follow a known Beta(d/2, d/2) distribution. Lloyd-Max scalar quantization: Quantize each ...
GitHub

Python implementation of the TurboQuant and QJL vector quantization algorithms.

turboquant-py implements the TurboQuant and QJL vector quantization algorithms from Google Research (ICLR 2026 / AISTATS 2026). It compresses high-dimensional floating-point vectors to 1-4 bits per ...
IEEE

A lossless compression method based on vector quantization and linear prediction for 3D GIIRS datacubes

Geostationary Interferometric Infrared Sounder (GIIRS, launched in 2016) [1], [2], the appearance of which is definitely a huge step in remote sensing and meteorological observation, is a Fourier ...
Wired

Senators Demand to Know How Much Energy Data Centers Use

Democratic senator Elizabeth Warren and Republican senator Josh Hawley are urging the US’s central energy information agency to provide better information on how much electricity data centers actually ...
SDxCentral

TurboQuant: Did Google just drop a compression algorithm capable of stemming RAMageddon?

Google thinks it's found the answer, and it doesn't require more or better hardware. Originally detailed in an April 2025 paper, TurboQuant is an advanced compression algorithm that’s going viral over ...
IEEE

Optimizing Contextual Speech Recognition Using Vector Quantization for Efficient Retrieval

Abstract: Neural contextual biasing allows speech recognition models to leverage contextually relevant information, leading to improved transcription accuracy. However, the biasing mechanism is ...