Get Started

This guide shows you how to set up a Genkit Java project and make your first AI generation call.

Prerequisites

• Java 21+
• Maven 3.6+
• An API key from your preferred model provider (e.g., OpenAI, Google AI)

Configure the Maven Repository

Genkit Java artifacts are published to GitHub Packages. Even though the packages are public, you need to configure the repository in your project and provide authentication.

1. Add the GitHub Packages repository

Add the following repository to your pom.xml:

<repositories>
    <repository>
        <id>github</id>
        <url>https://maven.pkg.github.com/genkit-ai/genkit-java</url>
        <snapshots>
            <enabled>true</enabled>
        </snapshots>
    </repository>
</repositories>

2. Configure authentication

GitHub Packages requires authentication even for public packages. Create or edit your ~/.m2/settings.xml file:

<settings>
    <servers>
        <server>
            <id>github</id>
            <username>YOUR_GITHUB_USERNAME</username>
            <password>YOUR_GITHUB_TOKEN</password>
        </server>
    </servers>
</settings>

Replace YOUR_GITHUB_USERNAME with your GitHub username and YOUR_GITHUB_TOKEN with a personal access token (classic) that has at least the read:packages scope.

Tip

You can also use environment variables to avoid hardcoding credentials:

<server>
    <id>github</id>
    <username>${env.GITHUB_ACTOR}</username>
    <password>${env.GITHUB_TOKEN}</password>
</server>

Installation

Add the following dependencies to your Maven pom.xml:

<!-- Core Genkit framework -->
<dependency>
    <groupId>com.google.genkit</groupId>
    <artifactId>genkit</artifactId>
    <version>1.0.0-SNAPSHOT</version>
</dependency>

<!-- Choose a model plugin (example: OpenAI) -->
<dependency>
    <groupId>com.google.genkit</groupId>
    <artifactId>genkit-plugin-openai</artifactId>
    <version>1.0.0-SNAPSHOT</version>
</dependency>

<!-- Choose a server plugin (example: Jetty) -->
<dependency>
    <groupId>com.google.genkit</groupId>
    <artifactId>genkit-plugin-jetty</artifactId>
    <version>1.0.0-SNAPSHOT</version>
</dependency>

Configure your API key

Set the environment variable for your model provider:

# For OpenAI
export OPENAI_API_KEY=your-api-key

# For Google GenAI (Gemini)
export GOOGLE_GENAI_API_KEY=your-api-key

# For Anthropic (Claude)
export ANTHROPIC_API_KEY=your-api-key

Create your first application

import com.google.genkit.Genkit;
import com.google.genkit.GenkitOptions;
import com.google.genkit.ai.GenerateOptions;
import com.google.genkit.ai.GenerationConfig;
import com.google.genkit.ai.ModelResponse;
import com.google.genkit.plugins.openai.OpenAIPlugin;
import com.google.genkit.plugins.jetty.JettyPlugin;
import com.google.genkit.plugins.jetty.JettyPluginOptions;

public class Main {
    public static void main(String[] args) throws Exception {
        // Create server plugin
        JettyPlugin jetty = new JettyPlugin(JettyPluginOptions.builder()
            .port(8080)
            .build());

        // Create Genkit with plugins
        Genkit genkit = Genkit.builder()
            .options(GenkitOptions.builder()
                .devMode(true)
                .reflectionPort(3100)
                .build())
            .plugin(OpenAIPlugin.create())
            .plugin(jetty)
            .build();

        // Generate text
        ModelResponse response = genkit.generate(
            GenerateOptions.builder()
                .model("openai/gpt-4o-mini")
                .prompt("Tell me a fun fact!")
                .config(GenerationConfig.builder()
                    .temperature(0.9)
                    .maxOutputTokens(200)
                    .build())
                .build());

        System.out.println(response.getText());

        // Start the HTTP server (blocks until stopped)
        jetty.start();
    }
}

Running with the Dev UI

Install the Genkit CLI and start with the developer UI:

# Install Genkit CLI
npm install -g genkit

# Start your app with Dev UI
genkit start -- mvn exec:java

The Dev UI launches at http://localhost:4000 where you can test flows, inspect traces, and debug your AI logic.

Using Google GenAI (Gemini) instead

Swap the plugin to use Gemini models:

<dependency>
    <groupId>com.google.genkit</groupId>
    <artifactId>genkit-plugin-google-genai</artifactId>
    <version>1.0.0-SNAPSHOT</version>
</dependency>

import com.google.genkit.plugins.googlegenai.GoogleGenAIPlugin;

Genkit genkit = Genkit.builder()
    .plugin(GoogleGenAIPlugin.create(System.getenv("GOOGLE_GENAI_API_KEY")))
    .build();

ModelResponse response = genkit.generate(
    GenerateOptions.builder()
        .model("googleai/gemini-2.0-flash")
        .prompt("Tell me a fun fact!")
        .build());

Next steps

Creating Flows Define observable AI workflows

Generating Content Learn the full model API

Tool Calling Let models call your functions

Dev UI & CLI Set up your dev workflow