aa

What if your web app could run generative AI without a server? See how Chrome AI uses Gemini Nano for powerful, client-side features like summarization and image analysis.

#1about 5 minutes

Introducing generative AI in the browser with Chrome AI

Chrome AI brings the Gemini Nano model directly into the browser for on-device processing, but requires handling a one-time model download.

#2about 12 minutes

A tour of the task-specific Chrome AI APIs

Explore the suite of built-in APIs for common tasks like translation, text summarization, rewriting, proofreading, and language detection.

#3about 7 minutes

The JavaScript implementation pattern for Chrome AI

Learn the essential three-step coding pattern for using any Chrome AI feature: check for existence, verify availability, and create an instance while handling model downloads.

#4about 5 minutes

Live demonstrations of text manipulation APIs

See live code examples of the translator, summarizer, and rewriter APIs in action, showcasing different options for output style and length.

#5about 5 minutes

Analyzing images with the general-purpose Prompt API

Discover how to use the flexible Prompt API for multimodal tasks like image analysis, and see how enriching prompts with external data like geolocation improves results.

#6about 2 minutes