GadgetBond

  • Latest
  • Features
  • How-to
  • Tech
    • AI
      • Apple Intelligence
      • Gemini AI
      • Google DeepMind
      • Anthropic
      • Claude AI
      • Claude Code
      • OpenAI
      • ChatGPT
      • Codex
      • Perplexity
      • SpaceXAI
      • Grok AI
      • Microsoft Copilot
      • Meta AI
    • Amazon
    • Apple
    • CES
    • Computing
    • Creators
    • Google
    • Meta
    • Microsoft
    • Mobile
    • NVIDIA
    • Samsung
    • Security
    • Smart Home
    • Sony
    • Xbox
    • YouTube
  • Transportation
    • Audi
    • BMW
    • Cadillac
    • E-Bike
    • Ferrari
    • Ford
    • Honda Prelude
    • Lamborghini
    • McLaren
    • Mercedes
    • Porsche
    • Rivian
    • Tesla
  • Culture
    • Gaming
    • Streaming
    • Apple TV
    • Disney
    • Hulu
    • Marvel
    • HBO Max
    • Netflix
    • Paramount
    • SHOWTIME
    • Spotify
    • Star Wars
Add GadgetBond as a preferred source to see more of our stories on Google.
Font ResizerAa
GadgetBondGadgetBond
  • Latest
  • Tech
  • AI
  • Features
  • Deals
  • How-to
  • Apps
  • Computing
  • Gaming
  • Mobile
  • Streaming
  • Transportation
Search
  • Latest
  • Deals
  • Buying Guide
  • How-to
  • Features
  • Tech
    • Amazon
    • Apple
    • CES
    • Computing
    • Creators
    • Google
    • Meta
    • Microsoft
    • Mobile
    • NVIDIA
    • Samsung
    • Security
    • Smart Home
    • Sony
    • Xbox
    • YouTube
  • AI
    • Apple Intelligence
    • Gemini AI
    • Google DeepMind
    • Anthropic
    • Claude AI
    • Claude Code
    • OpenAI
    • ChatGPT
    • Codex
    • Perplexity
    • SpaceXAI
    • Grok AI
    • Microsoft Copilot
    • Meta AI
  • Transportation
    • Audi
    • BMW
    • Cadillac
    • E-Bike
    • Ferrari
    • Ford
    • Lamborghini
    • McLaren
    • Mercedes
    • Porsche
    • Rivian
    • Tesla
  • Culture
    • Gaming
    • Streaming
    • Apple TV
    • Disney
    • Hulu
    • Marvel
    • HBO Max
    • Netflix
    • Paramount
    • SHOWTIME
    • Spotify
    • Star Wars
Follow US
AIGoogleTech

Google’s Gemma 3n AI can run locally on smartphones with 2GB RAM

Gemma 3n is a lightweight but capable AI model from Google that can be deployed locally for private, fast, and on-device processing.

By
Shubham Sawarkar
Shubham Sawarkar's avatar
ByShubham Sawarkar
Editor-in-Chief
Shubham Sawarkar is a technology journalist at GadgetBond, covering Apple, AI, consumer technology, software, gadgets, and product launches. His work explores emerging technologies, industry developments, product...
Follow:
- Editor-in-Chief
Jun 27, 2025, 7:51 AM EDT
Share
A logo of the Google Gemma AI model.
Image: Google
SHARE

When Google first unveiled its Gemma 3 family of AI models earlier this year, it hinted at a future where powerful language understanding wouldn’t be confined to data centers. On Thursday, the company delivered on that promise by releasing Gemma 3n—a fully open-source model refined for on-device use and capable of running on as little as 2GB of RAM. This means developers can deploy a sophisticated large language model (LLM) directly on smartphones and other low-resource hardware, making AI more accessible, private, and responsive than ever before.

In May, Google teased the “Nano” aspirations of Gemma 3n at its I/O developer conference, promising an AI small enough to fit in your pocket. The latest release confirms those ambitions: despite a raw parameter count of 5 billion (E2B) and 8 billion (E4B), Gemma 3n behaves like a 2 billion or 4 billion-parameter model in terms of memory footprint—just 2GB and 3GB of RAM, respectively. By offloading less-critical weights to “extra layer embeddings” handled by the CPU, Gemma 3n keeps its active memory lean, ensuring fast, local inference even on modest hardware.

At the heart of Gemma 3n lies Google’s Matryoshka Transformer, or MatFormer. Inspired by Russian nesting dolls, this architecture trains a larger model (E4B) alongside a nested smaller one (E2B), sharing weights where it counts and dynamically switching between the two during inference. The result is what Google calls a “mobile-first architecture,” where a single model package serves both high-performance and lightweight use cases without compromising quality.

A critical innovation enabling Gemma 3n’s efficiency is Per-Layer Embeddings (PLE). Traditional transformers load all parameters into GPU memory, but PLE keeps only the most essential ones in fast memory (VRAM), shuffling the rest in and out via the CPU. Coupled with activation quantization and key-value-cache (KVC) sharing, this approach slashes RAM requirements and speeds up response times—on mobile devices, Gemma 3n is roughly 1.5× faster than its predecessor, Gemma 3 4B, while delivering superior output quality.

Gemma 3n isn’t just a text model—it’s multimodal. It natively ingests images, audio, video, and text, though it currently generates text only. It’s also a global citizen, supporting 140 languages for text inputs and 35 languages when working with multimodal data, making it a versatile tool for developers worldwide. Best of all, Google has open-sourced both the model weights and a “cookbook” of recipe-style instructions for fine-tuning and deployment under a permissive Gemma license, allowing academic researchers and commercial teams alike to innovate without legal headaches.

Not content to ship one-size-fits-all models, Google is also releasing MatFormer Lab—a toolkit that lets developers experiment with different nesting depths, parameter allocations, and quantization settings to craft custom-sized Gemma derivatives. Whether you need a hyper-lean 1 B-equivalent model for a microcontroller or a beefier 6 B-equivalent engine for a laptop, MatFormer Lab provides a playground to tune for specific latency, memory, and accuracy trade-offs.

You don’t need a PhD to start tinkering with Gemma 3n. The full model line is available on Hugging Face and via Google’s Kaggle listings. For a zero-install trial, head to Google AI Studio, where you can spin up inference jobs or even deploy directly to Cloud Run. Want to see it in action? Google provides sample notebooks, Docker images, and command-line scripts to help you integrate Gemma 3n into chatbots, virtual assistants, or edge-AI demos within minutes.


Discover more from GadgetBond

Subscribe to get the latest posts sent to your email.

Topic:Gemini AI (formerly Bard)Google DeepMind

Most Popular

Four AI apps to act as your personal life admin team
Marvel Comics app is finally here, bringing 35,000+ comics and a new reading experience
Apple announces new regional taxes and December price adjustments for apps and in-app purchases
Google introduces Stitch Loop for continuous AI-powered code optimization
How these three Apple-recommended apps turn everyday stress into quiet me-time

Also Read

Hot air balloons float over snowy mountains under a bright sun, with "Genie 4" centered and on-screen WASD, SPACE, and arrow controls at the bottom.

Google Project Genie gets a major upgrade with Genie 4

Google Gemini Live interface displayed on a smartphone, showing the greeting “Ready when you are, Camila!” with voice, camera, and screen-sharing controls.

Google brings the Gemini app to lower-memory Android (Go edition) devices

Apple App Store icon glowing against a blue background, featured in the 2026 App Store Awards announcement.

Apple reveals the 45 finalists for the 2026 App Store Awards

Keke Palmer holds the Apple Podcasts icon while smiling against a pink and purple gradient background.

Apple Podcasts names Keke Palmer’s show its 2026 Show of the Year

Google Search AI Mode temporary thread interface displaying a “Welcome, stranger” message and an input box, with a notice explaining that temporary threads won't appear in Search history or be used to train Google's generative AI models.

Google adds temporary threads to AI Mode in Search for off-the-record queries

Google Wallet for iOS displayed on an iPhone, featuring a digital boarding pass, Google Store delivery notification, loyalty cards, membership cards, concert tickets, and other passes. The accompanying text reads, “Google Wallet on iOS — Now available on iOS.”

Google Wallet arrives on iPhone in the U.S. for tickets, passes, and agentic purchases

Gemini interface showing "What can I get done for you?" above a "Describe a task" search bar set to "Hydrogen" and a "Task" toggle selected.

Google Gemini can now handle everyday tasks for you

Google Vids logo

Google Vids now lets anyone with a personal Google account create custom digital avatars for free

Company Info
  • Homepage
  • Support my work
  • Latest stories
  • Company updates
  • GDB Recommends
  • Daily newsletters
  • About us
  • Contact us
  • Write for GadgetBond
  • Editorial guidelines
Legal
  • Privacy Policy
  • Cookies Policy
  • Terms & Conditions
  • DMCA
  • Disclaimer
  • Accessibility Policy
  • Security Policy
  • Do Not Sell or Share My Personal Information
Socials
Follow US

Disclosure: We love the products we feature and hope you’ll love them too. If you purchase through a link on our site, we may receive compensation at no additional cost to you. Read our ethics statement. Please note that pricing and availability are subject to change.

Copyright © 2026 GadgetBond. All Rights Reserved. Use of this site constitutes acceptance of our Terms of Use and Privacy Policy | Do Not Sell/Share My Personal Information.