IBM details 0.7-nanometer chip design aimed at AI-era computing
What's new: IBM said the prototype packs nearly 100 billion transistors and could reach production within five years.
IBM detailed a new semiconductor design it says can push chipmaking below 1 nanometer, a milestone the company is pitching as a path to faster and more efficient AI computing. The prototype uses a 0.7-nanometer transistor architecture called NanoStack, which stacks transistors vertically to fit more computing power into the same space. IBM said the design places nearly 100 billion transistors on a fingernail-sized chip, roughly double the density of its 2-nanometer technology introduced in 2021. In tests, the company said the prototype delivered up to 50% better performance or 70% greater energy efficiency. IBM said commercial production remains several years away and has not named a manufacturing partner.