/spring-ai-openrouter-2025

Spring AI with OpenRouter to demo Gemini 2.5 Pro and LLama 4 Maverick

Primary LanguageJava

Spring AI - OpenRouter - Google Gemini 2.5 Pro & Llama 4 Maverick

A Spring Boot application demonstrating integration with multiple AI models through OpenRouter API using Spring AI framework.

Overview

This project showcases how to build a simple chat application that can:

  • Connect to various AI models via OpenRouter
  • Process and handle multimedia files in prompts
  • Provide a RESTful API endpoint for interacting with AI models

Technologies Used

  • Java 21
  • Spring Boot 3.4.4
  • Spring AI 1.0.0-M6
  • OpenRouter API (providing access to models like Meta Llama 4 Maverick and Google Gemini 2.5 Pro)

Project Structure

└── src
    ├── main
    │   ├── java
    │   │   └── vn/cloud/spring_ai_openrouter_demo
    │   │       ├── ChatClientConfiguration.java - Configures the ChatClient bean
    │   │       ├── ChatController.java - REST API endpoint for chat interactions
    │   │       ├── JacksonConfiguration.java - Jackson JSON parser configuration
    │   │       └── SpringAiOpenrouterDemoApplication.java - Main application class
    │   └── resources
    │       └── application.yml - Application configuration properties
    └── test
        └── java
            └── vn/cloud/spring_ai_openrouter_demo
                └── SpringAiOpenrouterDemoApplicationTests.java - Basic context load test

Features

  • REST API Endpoint: Send questions and files to AI models via a POST request
  • Multi-Model Support: Configure different AI models through application.yml
  • Multimedia Support: Upload and process files alongside text prompts
  • High File Size Limits: Configured to handle files up to 20MB

Setup and Configuration

Prerequisites

  • Java 21 or higher
  • Gradle 8.x
  • OpenRouter API key

Configuration

  1. Open src/main/resources/application.yml
  2. Replace { OPEN_ROUTER_API_KEY } with your actual OpenRouter API key
  3. Optionally change the model configuration: 
    model: google/gemini-2.5-pro-exp-03-25:free
# or uncomment to use:
# model: meta-llama/llama-4-maverick:free

Building the Project

./gradlew clean build

Running the Application

./gradlew bootRun

Or after building:

java -jar build/libs/spring-ai-openrouter-demo-0.0.1-SNAPSHOT.jar

API Usage

Chat Endpoint

Send a POST request to /chat with:

  • question: The text prompt to send to the AI model
  • files: Optional files to include with the prompt (supports multiple files)

Example using curl:

curl -X POST http://localhost:8080/chat \
  -F "question=Describe what you see in this image" \
  -F "files=@/path/to/your/image.jpg"

Example using Postman:

  1. Set request type to POST
  2. URL: http://localhost:8080/chat
  3. Body: form-data
    • Key: question, Value: "Describe what you see in this image"
    • Key: files, Value: [select file], Type: File

Development

This project uses standard Spring Boot development practices. You can import it into any IDE that supports Gradle projects.

License

This project is open source and available under the MIT License.

Additional Resources