How to Build a Spring AI Hello World with Ollama and DeepSeek Locally

Overview

Spring AI 1.0 has been released; this guide demonstrates how to build a simple "Hello World" AI application by installing Ollama, pulling the DeepSeek‑R1 model, creating a Spring Boot project, writing a ChatController, and testing the interaction.

Step 1: Install Ollama

Download Ollama from https://ollama.com/download, install it to a custom directory (e.g., D:\Ollama), and set the model directory environment variable: setx OLLAMA_MODELS "D:\Ollama\models" /M Verify the installation with ollama -v, which should output version 0.5.11.

Step 2: Install DeepSeek

From the Ollama site, pull the lightweight 1.5b DeepSeek‑R1 model (1.5 billion parameters): ollama pull deepseek-r1:1.5b Run the model locally with:

ollama run deepseek-r1:1.5b

Step 3: Create Spring AI Project

Generate a Spring Boot project with Spring Initializr (Java 17) and replace the default OpenAI starter with the Ollama starter in pom.xml:

<dependency>
    <groupId>org.springframework.ai</groupId>
    <artifactId>spring-ai-starter-model-ollama</artifactId>
</dependency>

Configure application.yml to point to the local Ollama server and the DeepSeek model:

spring:
  http:
    encoding:
      charset: UTF-8
      enable: true
      force: true
  ai:
    ollama:
      base-url: http://localhost:11434
      chat:
        model: deepseek-r1:1.5b

Step 4: Implement Chat Controller

Use Spring AI’s ChatClient to forward user messages to the model:

package com.myai.demo.controller;

import org.springframework.ai.chat.client.ChatClient;
import org.springframework.web.bind.annotation.*;

@RestController
@RequestMapping("/ai")
public class ChatController {

    private final ChatClient chatClient;

    public ChatController(ChatClient.Builder chatClient) {
        this.chatClient = chatClient.build();
    }

    @GetMapping("/chat")
    public String chat(@RequestParam("message") String message) {
        try {
            return chatClient.prompt().user(message).call().content();
        } catch (Exception e) {
            return "Exception";
        }
    }
}

Step 5: Test and Summary

Run the Spring Boot application and send a request to /ai/chat?message=hello. The model replies with a friendly greeting, confirming that the "Hello World" AI integration works.

This tutorial shows that a Java backend can interact with a locally hosted large language model, opening many possibilities for AI‑enhanced applications.