ainewsblitz.com

Breaking

Google Reportedly Developing 'Frozen v2' Chip to Run Gemini More Efficiently

Google Reportedly Developing 'Frozen v2' Chip to Run Gemini More Efficiently
  • Infra & Chips
  • Foundation Models

Google is developing a specialized AI server chip, internally dubbed "Frozen v2," that would etch parts of its Gemini model architecture directly into silicon in a bid to dramatically improve inference efficiency, according to a report from The Information. The chip is said to be distinct from Google's existing Tensor Processing Units and is aimed at running Gemini far more cheaply and with far less power.

Continue reading

The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.

$20
Read this article
$29/month
Unlimited — all 27,934 articles, the full archive, and comprehension quizzes
Save 72%
$98/year
≈ $8.17/month
Unlimited, billed once a year