ainewsblitz.com

Breaking

Google Develops 'Frozen v2' Chip That Hardwires Gemini to Cut AI Inference Costs

Google Develops 'Frozen v2' Chip That Hardwires Gemini to Cut AI Inference Costs
  • Infra & Chips
  • Foundation Models

Google is building a specialized server chip, reportedly code-named "Frozen v2," that bakes elements of its Gemini AI models directly into silicon in an effort to dramatically lower the cost of running those models. The chip is designed to complement rather than replace Google's Tensor Processing Units, and it is aimed at squeezing far more efficiency out of the inference stage, when a trained model generates outputs for users.

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