aiqbench takeChips & Infra

Google is developing a custom server chip codenamed "Frozen v2" aimed at significantly improving the energy efficiency of its Gemini models by 6x to 10x in terms of tokens per watt, with an expected release around 2028. This strategic move underscores the industry-wide shift toward co-designing hardware and software to reduce dependency on Nvidia while addressing growing concerns over AI compute costs and capacity constraints.

Sources coveringTechCrunch, LessWrong, Effective Altruism Forum, AI Alignment Forum

What this isA model's reading of the article and related coverage, written for aiqbench. Not the publisher's words.

Google is working on a new AI chip designed to make Gemini more efficient

TechCrunchJuly 20, 20262 min read

Alphabet, Google's parent company, is reportedly working on a new chip designed to make its Gemini models run much more efficiently.

Alphabet, Google’s parent company, is designing a new server chip to help its in-house Gemini models operate more efficiently. The new chip, internally dubbed “Frozen v2,” is slated to be released sometime in 2028, The Information reported, citing anonymous sources. According…

Excerpt shown under the publisher's RSS terms · 41 words · full text and figures are on the publisher's site

Read at TechCrunch ↗Ask about this article