Skip to main content
Gemini 3.1 Pro has been released! Explore the latest multimodal capabilities and enhanced performance.

​
Welcome to Generative AI on Google Cloud

This comprehensive repository contains notebooks, code samples, sample applications, and resources demonstrating how to use, develop, and manage generative AI workflows using Generative AI on Google Cloud with Vertex AI. With over 350 Jupyter notebooks and real-world applications, you’ll learn how to leverage Google’s most powerful AI models including Gemini, Imagen, and specialized services like Vertex AI Search.

​
Key Features

Gemini Multimodal Models

Build with state-of-the-art multimodal AI that understands text, images, video, and audio

AI Agents

Create intelligent agents with the Agent Development Kit and Agent Engine

RAG & Search

Ground your AI with retrieval augmented generation and Vertex AI Search

Vector Embeddings

Build semantic search with text and multimodal embeddings

Image Generation

Generate and edit images with Imagen on Vertex AI

Model Evaluation

Evaluate and compare models with the Gen AI Evaluation SDK

​
Quick Start

Get up and running in minutes:
1

Set up Google Cloud

Configure your Google Cloud project and enable Vertex AI APIsEnvironment Setup Guide
2

Install the SDK

Install the Vertex AI Python SDK and authenticate
pip install google-cloud-aiplatform
3

Run your first notebook

Start with a simple Gemini exampleQuickstart Guide

​
What You’ll Find Here

​
Gemini Models

Discover Gemini through starter notebooks, use cases, function calling, multimodal processing, and sample applications. Learn about:
  • Multimodal capabilities - Process text, images, video, and audio in a single model
  • Function calling - Connect Gemini to external tools and APIs
  • Grounding - Ground responses with Google Search or your own data
  • Context caching - Optimize costs by caching large contexts
  • Code execution - Run Python code directly within Gemini

​
AI Agents

Build stateful, context-aware conversational agents using:
  • Agent Engine - Managed service for deploying production agents
  • Agent Development Kit (ADK) - Framework for building custom agents
  • Memory Bank - Persistent memory for conversational context
Implement retrieval augmented generation with:
  • RAG Engine - Managed RAG service on Vertex AI
  • Vertex AI Search - Enterprise search across your data
  • Grounding techniques - Connect models to authoritative sources

​
Production Features

Take your applications to production with:
  • Model evaluation - Comprehensive testing and benchmarking
  • Orchestration - LangChain, LlamaIndex, and custom workflows
  • Responsible AI - Safety filters and content moderation
  • Open models - Deploy and fine-tune open source models from Model Garden

​
Repository Structure

Core Gemini model samples including getting started guides, function calling, multimodal processing, evaluation, and sample applications
Text and multimodal embeddings with Vector Search integration
Image generation, editing, and visual question answering with Imagen
Speech recognition and audio processing with Chirp
Agent development samples including ADK and Agent Engine

​
Community & Support

GitHub Repository

View source code, report issues, and contribute

Contributing Guide

Learn how to contribute to this repository

Learning Resources

Explore Google Cloud’s Generative AI documentation

Vertex AI Documentation

Official Vertex AI product documentation
This repository is for demonstrative purposes only and is not an officially supported Google product. For production applications, refer to the official Vertex AI documentation.

​
Next Steps

Setup

Configure your environment

Quickstart

Run your first example

Explore Use Cases

Browse real-world applications

Build docs developers (and LLMs) love

Get started for freeTalk to us