The Gemma 3n model has been fully released, building on the success of previous Gemma models and bringing advanced on-device multimodal capabilities to edge devices with unprecedented performance. Explore Gemma 3n's innovations, including its mobile-first architecture, MatFormer technology, Per-Layer Embeddings, KV Cache Sharing, and new audio and MobileNet-V5 vision encoders, and how developers can start building with it today.
| CARVIEW |
Select Language
HTTP/2 301
content-type: text/html; charset=utf-8
location: https://googledevelopers.blogspot.com
cache-control: private, max-age=2592000
pragma:
vary: Cookie
vary: Accept-Encoding
content-security-policy: base-uri 'self'; object-src 'none'; script-src 'strict-dynamic' 'unsafe-inline' https: http: 'nonce-ShLQUUqRHyd1vB5oYhPfiLjmOzUn0r' 'unsafe-eval'; frame-ancestors 'self' https://code-assist-free-tier-autopush.corp.google.com https://code-assist-free-tier-staging.corp.google.com https://code-assist-free-tier.corp.google.com https://*.proxy.googlers.com; report-uri https://csp.withgoogle.com/csp/devsite/v2
strict-transport-security: max-age=63072000; includeSubdomains; preload
x-xss-protection: 0
x-content-type-options: nosniff
expires: 0
content-encoding: gzip
x-cloud-trace-context: 2f42b755ca4bdc2322328ed03c2055bf
date: Sat, 19 Jul 2025 02:37:25 GMT
server: Google Frontend
content-length: 185
alt-svc: h3=":443"; ma=2592000,h3-29=":443"; ma=2592000
HTTP/2 301
content-type: text/html; charset=UTF-8
location: https://developers.googleblog.com/
content-encoding: gzip
date: Sat, 19 Jul 2025 02:37:27 GMT
expires: Sat, 19 Jul 2025 02:37:27 GMT
cache-control: private, max-age=0
x-content-type-options: nosniff
x-xss-protection: 1; mode=block
content-length: 198
server: GSE
alt-svc: h3=":443"; ma=2592000,h3-29=":443"; ma=2592000
HTTP/2 302
content-type: text/html; charset=utf-8
location: /en/
vary: Accept-Language, Cookie
x-cloud-trace-context: 5d43b696f9c5c453afb5636513cbe15e
date: Sat, 19 Jul 2025 02:37:27 GMT
server: Google Frontend
content-length: 0
alt-svc: h3=":443"; ma=2592000,h3-29=":443"; ma=2592000
HTTP/2 200
content-type: text/html; charset=utf-8
vary: Accept-Encoding
vary: Cookie
x-frame-options: DENY
strict-transport-security: max-age=2592000; includeSubdomains
x-content-type-options: nosniff
x-xss-protection: 1; mode=block
content-security-policy: base-uri 'self'; object-src 'none'; script-src 'strict-dynamic' http: https: 'nonce-5vC1yeo3ivriAZ327SHHFw==' *.google-analytics.com *.googletagmanager.com *.gstatic.com *.youtube.com *.googleapis.com
referrer-policy: same-origin
cross-origin-opener-policy: same-origin
content-language: en
content-encoding: gzip
x-cloud-trace-context: 08e526416616ca58afb5636513cbe722
date: Sat, 19 Jul 2025 02:37:29 GMT
server: Google Frontend
content-length: 15017
alt-svc: h3=":443"; ma=2592000,h3-29=":443"; ma=2592000
Home
- Google Developers Blog
Featured articles
Introducing Gemma 3n: The developer guide
Gemini
Google Labs
AI
AI Homepage
Announcements
From idea to app: Introducing Stitch, a new way to design UIs
Stitch, a new Google Labs experiment, uses AI to generate UI designs and frontend code from text prompts and images, aiming to streamline the design and development workflow, offering features like UI generation from natural language or images, rapid iteration, and seamless paste to Figma and front-end code.
Gemini
Google AI Studio
AI
AI Homepage
Announcements
Imagen 4 is now available in the Gemini API and Google AI Studio
Imagen 4, Google's advanced text-to-image model, is now available in paid preview via the Gemini API and Google AI Studio, offering significant quality improvements, especially for text generation within images. The Imagen 4 family includes Imagen 4 for general tasks and Imagen 4 Ultra for high-precision prompt adherence, with all generated images featuring a non-visible SynthID watermark.
AI Edge
Mobile
Web
AI
Mobile Homepage
Announcements
Industry Trends
On-device small language models with multimodality, RAG, and Function Calling
Google AI Edge advancements, include new Gemma 3 models, broader model support, and features like on-device RAG and Function Calling to enhance on-device generative AI capabilities.
Android
Firebase
Gemini
Google AI Studio
Web
AI
Web Homepage
Mobile
Cloud
Announcements
What you should know from the Google I/O 2025 Developer keynote
Top announcements from Google I/O 2025 focus on building across Google platforms and innovating with AI models from Google DeepMind, with key focus on new tools, APIs, and features designed to enhance developer productivity and create AI-powered experiences using Gemini, Android, Firebase, and web.
Latest blogs
AI
See postsBuild with Veo 3, now available in the Gemini API
Simplify your Agent "vibe building" flow with ADK and Gemini CLI
Stanford’s Marin foundation model: The first fully open model developed using JAX
Gemini Embedding now generally available in the Gemini API
Announcing GenAI Processors: Build powerful and flexible Gemini applications
T5Gemma: A new collection of encoder-decoder Gemma models
Batch Mode in the Gemini API: Process more for less
Introducing Gemma 3n: The developer guide
Simulating a neural operating system with Gemini 2.5 Flash-Lite
Mobile
See postsAdding support for Google Pay within Android WebView
New Google Pay features to enhance your payment flows
On-device small language models with multimodality, RAG, and Function Calling
LiteRT: Maximum performance, simplified
Celebrating Flutter’s “Production Era”
Learn to build and run AI powered apps at Firebase Demo Day ‘24
TensorFlow Lite is now LiteRT
Indie Games Fund: Google Play’s $2m fund in Latin America is back
Streamlining LLM Inference at the Edge with TFLite
Web
See postsGoogle Pay inside sandboxed iframe for PCI DSS v4 compliance
Bringing Gemini intelligence to Google Home APIs
Explore the latest updates on Google Wallet
What you should know from the Google I/O 2025 Developer keynote
Build and train a recommender system in 10 minutes using Keras and JAX
How It’s Made: Little Language Lessons uses Gemini’s multilingual capabilities to p…
Usability and safety updates to Google Auth Platform
Building a better smart home
Unlocking the Potential of Quantum Computing
Cloud
See postsUnlock Gemini’s reasoning: A step-by-step guide to logprobs on Vertex AI
Enterprise truth in action: Apigee API hub fueling powerful Developer Portals
Advancing agentic AI development with Firebase Studio
Unlock deeper insights with the new Python client library for Data Commons
Google Cloud donates A2A to Linux Foundation
Data-driven marketing starts with developers
What's new with Agents: ADK, Agent Engine, and A2A Enhancements
Google Cloud announces general availability of APIM Operator for Apigee
Announcing the Agent2Agent Protocol (A2A)
Follow Google for Developers
Subscribe to join a community of creative developers and learn the latest in Google technology.
Learn more
Subscribe to Google for Developers news. Your information will be used in accordance with Google’s privacy policy.
Subscribe