GBAF Logo
Global Banking & Finance Awards® 2026 Nominations open, free to enter Nominate now →
Google plans new chip to run Gemini models more efficiently, the Information reports - Finance news and analysis from Global Banking & Finance Review
Finance

Google plans new chip to run Gemini models more efficiently, the Information reports

Published by Global Banking & Finance Review

Posted on July 20, 2026

2 min read

· Last updated: July 20, 2026

Add as preferred source on Google

Google plans new chip to run Gemini models more efficiently, the Information reports

Google's Development of the "Frozen v2" AI Server Chip

July 20 (Reuters) - Google is developing a new server chip that would incorporate elements of its Gemini model directly into the hardware, in a bid to serve its AI models more efficiently to users, the Information reported on Monday, citing people familiar with the matter.

The Alphabet-owned company expects the new chip, informally dubbed "Frozen v2," to help address an AI computing capacity crunch that has fueled internal tensions and prompted Google Cloud to decline deals with outside customers, the report said.

Shares of Alphabet were up 3.3% in early trading. 

Key Details of the "Frozen v2" Chip

Here are some details:

Deployment Timeline and Design

• Google plans to deploy the chip as soon as 2028, though engineers are still finalizing its design and the amount of model information that will be hardwired, the report said.

Efficiency Improvements

• The chip could be six to 10 times more efficient than Google's latest custom AI chips based on the number of AI tokens served per unit of power, according to the report.

Google Cloud's Approach

• "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," a Google Cloud spokesperson said.

Project Scope and Differentiation

• The 'Frozen' project is aimed at creating a new set of homegrown chips apart from Google's tensor processing units (TPUs), rather than replacing them, the report said.

Recent Developments in Gemini AI

• Bloomberg News reported last week that Google delayed the launch of its latest Gemini AI model after it fell short of internal goals, with the company working to improve its capabilities, particularly in ‌coding.

Reporting and Editorial Credits

(Reporting by Rashika Singh in Bengaluru; Editing by Leroy Leo)

Key Takeaways

  • Google aims to alleviate internal AI compute shortages and external customer capacity constraints by creating Frozen v2, embedding model components in the chip.
  • The new chip could deliver 6–10× higher efficiency in AI tokens served per unit of power compared to current custom AI chips.
  • Frozen v2 is intended to complement—not replace—Google’s existing TPUs, helping address cloud demand backlogs and internal tensions over compute allocation.

Frequently Asked Questions

What is Google's new server chip project called?
Google's new server chip project is informally dubbed 'Frozen v2'.
When does Google plan to deploy the Frozen v2 chip?
Google plans to deploy the Frozen v2 chip as soon as 2028, though its design is still being finalized.
How much more efficient is Frozen v2 expected to be compared to other AI chips?
The Frozen v2 chip could be six to ten times more efficient than Google’s latest custom AI chips based on AI tokens served per unit of power.
Is Frozen v2 intended to replace Google’s TPUs?
No, the Frozen project aims to create a new set of homegrown chips apart from Google’s tensor processing units (TPUs), not replace them.
Why is Google developing this new chip for AI models?
Google is developing the chip to address AI computing capacity shortages and improve how efficiently it serves Gemini AI models to users.

Tags

Related Articles

More from Finance

Explore more articles in the Finance category