Google Designs a Gemini-Specific AI Chip Called Frozen v2
Google is designing Frozen v2, a chip that hardwires parts of its Gemini model into silicon and could serve six to ten times more tokens per watt, targeting deployment by 2028.
Explore AIstify's latest reporting, research, and expert analysis tagged with "tpu", collected in one continuously updated archive.
Google is designing Frozen v2, a chip that hardwires parts of its Gemini model into silicon and could serve six to ten times more tokens per watt, targeting deployment by 2028.
Google and Blackstone have formed a new AI infrastructure company focused on offering cloud access to Google’s TPU chips. The venture aims to expand alternatives to Nvidia-powered AI infrastructure with $5 billion in initial funding.
Anthropic has signed a compute agreement with SpaceX that adds access to more than 220,000 NVIDIA GPUs at the Colossus 1 data center. The added capacity is already being used to raise Claude usage limits and API availability.
Broadcom is expanding its role in AI infrastructure through new chip and compute deals with Google and Anthropic. The move reflects accelerating demand for large-scale AI capacity.
AI chips are specialized processors designed to run artificial intelligence models efficiently. This guide explains what AI chips are, how they work, and why they are essential for modern AI systems.
Ricursive Intelligence hits a $4 billion valuation after raising $300 million to develop AI systems that autonomously design and improve AI chips.
Alphabet reached a $4 trillion market valuation after a 65% stock gain in 2025, fueled by AI investments including Gemini 3 and custom AI chips for enterprise and consumer products.
Nvidia asserts it is “a generation ahead” in AI chips after reports that Google may supply Meta with its processors, spurring market competition.
Google executives told employees the company must double its compute capacity every six months to keep pace with surging demand for AI services, underscoring the rising cost and competition in cloud infrastructure.
The hardware, software, and cloud systems that power AI development, enabling data processing, model training, and large-scale deployment across industries.