Tags: large language models

Google Research unveiled TurboQuant, a new compression algorithm designed to dramatically reduce the memory footprint of large language models (LLMs) while also increasing inference speed. By targeting the key‑value cache—often described as a digital cheat sheet—TurboQuant can cut memory usage by up to six times and deliver performance gains of around eight times without sacrificing model quality. The technique relies on a novel PolarQuant conversion that represents vectors in polar coordinates, preserving essential information while enabling aggressive compression. Read more

Apple co‑founder Steve Wozniak told Fox Business that he is “not a fan” of artificial‑intelligence systems such as ChatGPT and Claude. He argued that AI lacks emotional depth, often provides overly detailed factual answers instead of storytelling, and can be unreliable. Wozniak’s comments contrast with Apple CEO Tim Cook’s optimism about AI and come as Apple pushes its own AI platform, Apple Intelligence, amid development challenges. Read more