FrontierThe story, in brief
Gemma 4 just replaced my whole local LLM stack - MakeUseOf
2.3B parameters. 1.5GB RAM. Google's Gemma 4 just made running cutting-edge AI locally viable for the first time.
The KeyNews take
Why it matters
Google's Gemma 4 represents a meaningful shift in on-device AI efficiency—a 2.3B model matching 70B-scale performance on minimal hardware removes a key barrier to local-first AI adoption and challenges the cloud-dependency thesis.
The key facts
6 to knowGemma 4: 2.3B parameters
Runs on 1.5GB RAM
Performance parity with 70B-scale models claimed
Offline/on-device capability on Pixel phones
Privacy-focused local execution
Published April 18, 2026
Go to the source
Reuters Technologynews.google.com
Publisher excerpt: Gemma 4 just replaced my whole local LLM stack MakeUseOf I replaced ChatGPT with Google's offline AI on my phone for 24 hours — here's my verdict Tom's Guide How Google’s 2.3B Gemma 4 Model Rivals 70B Giants on Just 1.5GB of RAM Geeky Gadgets Forget Gemini and Claude, this is the free game-changing…