Back to AIBriefs
LaunchAI ModelsDevelopers

Google releases Gemma 4 QAT models for mobile efficiency

Google releases Gemma 4 QAT models for mobile efficiencyblog.google

New QAT checkpoints reduce Gemma 4 E2B memory to 1GB. Available in Q4_0 and mobile-optimized formats for local inference on edge devices.

··Discuss
12 days ago
Google releases Gemma 4 QAT models for mobile efficiency — AIBriefs