Google is developing a new AI chip to make Gemini faster and more efficient

Google is developing a new custom server chip designed to run its Gemini artificial intelligence models more efficiently, according to a report by The Information, as the tech giant looks to ease growing demand for AI computing power.

The report, citing people familiar with the project, says the new processor—internally known as "Frozen v2"—would embed parts of Google's Gemini AI model directly into the chip's hardware, allowing AI tasks to be processed more efficiently while reducing power consumption, News.Az reports, citing Reuters.

The chip is intended to help Google overcome increasing pressure on its AI infrastructure, as demand for Gemini-powered services continues to rise. According to the report, limited computing capacity has even forced Google Cloud to turn down some potential customer deals.

Google reportedly aims to begin deploying the chip as early as 2028, although engineers are still refining the design and determining how much of the Gemini model will be integrated directly into the hardware.

If successful, the new chip could be six to ten times more efficient than Google's latest custom AI processors when measured by the number of AI-generated tokens produced per unit of power, the report said.

A Google Cloud spokesperson did not comment directly on the reported project but said the company is continuously exploring new technologies.

"Our teams are constantly researching and experimenting with new innovations. By co-designing our hardware and software from the ground up, we ensure our systems are integrated and highly optimized," the spokesperson said.

According to The Information, the "Frozen" project is intended to complement—not replace—Google's existing Tensor Processing Units (TPUs), which currently power much of the company's AI infrastructure.

The report comes days after Bloomberg News said Google had delayed the release of its latest Gemini AI model because it did not meet internal performance targets, particularly in coding capabilities, as engineers continued refining the system.

Shares of Alphabet, Google's parent company, rose more than 3% in early trading following the report.