Alphabet Developing New AI Chip for Gemini Models
Alphabet is developing an in-house server chip, 'Frozen v2,' to improve the efficiency of its Gemini AI models by up to ten times compared to current chips. The chip is expected to be released around 2028 and reflects a broader trend among tech firms to reduce reliance on third-party hardware like Nvidia's offerings.
Topics: Alphabet custom server chips Frozen v2 chip Gemini AI models Google Nvidia dependency