hotAI

2 min read

Google’s next AI chip could cut Gemini power use

Google is reportedly building a new server chip, Frozen v2, that could make Gemini six to 10 times more efficient by 2028.

Image: TechCrunch

Google is reportedly developing a new server chip aimed at making its Gemini models far more efficient. The chip, internally called “Frozen v2,” is expected to arrive in 2028, according to The Information, which cited anonymous sources.

The report says Frozen v2 could be six to 10 times more efficient than Google’s current AI chips, based on the number of tokens generated per unit of power. Google did not confirm the details when asked by TechCrunch, but it also did not deny them.

“Our teams are constantly researching and experimenting with new innovations to deliver maximum performance and efficiency for our users and customers. While not every project moves into production, this rigorous exploration is central to our full stack approach. By co-designing our hardware and software from the ground up, we ensure our systems are integrated and highly optimized for real-world workloads.”

Google

The effort fits a broader push across the industry. AI companies are increasingly designing their own chips both to improve efficiency and to reduce exposure to global shortages in AI compute. Cost and power use have become central concerns as investor enthusiasm around AI gives way to tougher scrutiny of spending.

There is also a strategic reason: major model makers are trying to reduce their dependence on Nvidia, which still dominates the AI chip market. In June, OpenAI announced its first custom inference chip, Jalapeño. Earlier this month, reports said Anthropic was discussing a new chip partnership with Samsung.

Recommended reading

South Korea to launch free domestic AI chatbot in 2026

That context matters for Alphabet, whose AI spending plans have drawn investor attention. Earlier this year, Google said it planned to spend between $180 billion and $190 billion to build out its AI strategy. The prospect of a much more efficient in-house chip appeared to reassure the market. After The Information report was published, Google’s stock rose about 3% on Monday morning, just ahead of the company’s earnings report later this week.

Ava Chen

AI Editor

Ava covers the rapidly evolving world of artificial intelligence, from foundational models and research labs to the real-world economics of intelligence. With a background in computational linguistics, she cuts through the hype to find out what actually works. She firmly believes that benchmarks are just marketing until reproduced in the wild.

via TechCrunch

// Keep reading