ainewsblitz.com

Breaking

Google's Gemma 4 26B Runs on a 13-Year-Old Xeon With No GPU, at Reading Speed

Google's Gemma 4 26B Runs on a 13-Year-Old Xeon With No GPU, at Reading Speed
  • Foundation Models
  • Open Source
  • Infra & Chips

A developer has run Google DeepMind's Gemma 4 26B A4B model on a 2016-era Intel Xeon server with no GPU at all, generating text at roughly 5 tokens per second — a pace comparable to human reading speed. The demonstration, shared in June 2026, has drawn attention as a striking example of how far efficient open models can stretch older, inexpensive hardware.

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