Google is developing a new server chip, codenamed "Frozen v2", that would embed the GEMINI model's design blueprint to deliver AI services more efficiently, sources said.
Shares of Alphabet jumped over 3% on the news.
The chip is intended to ease severe compute shortages that have triggered internal resource competition and forced Google Cloud to turn away some external customers. Project participants estimate Frozen v2 could process 6–10x more tokens per unit of power than Google's current latest in-house AI chip; engineers are still finalizing its core functions and component integration. Deployment is planned as early as 2028.