Google is Developing New Chip To Make Gemini More Efficient: Report

Google is developing a new server chip that will bring certain components of its Gemini AI model directly into the hardware, making its AI systems significantly more efficient to users, according to a report by The Information.

The report said the chip, codenamed "Frozen v2," is expected to help ease Google's AI computing capacity shortage, which has reportedly forced Google Cloud to decline deals with external customers.

Google is aiming to deploy the chip by 2028, with engineers still finalising its design and the amount of information that will be hardwired. According to the report, the chip could be six to ten times more efficient than Google's existing AI chips, based on the number of AI tokens processed per unit of power.

"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 optimised," a Google Cloud spokesperson said.

According to the report, the "Frozen" project is aimed at creating a new family of AI chips alongside Google's Tensor Processing Units (TPUs), rather than replacing it.

Last week, Bloomberg reported that Google had delayed the launch of its flagship AI model, Gemini 3.5 Pro, after it fell short of internal goals. The company is reportedly taking additional time to improve the model's capabilities, particularly in coding.