Integrating DeepSeek Large Language Model with Spring Boot to Build an AI Chat Application

In the digital age AI is pervasive, and Spring Boot remains a popular Java framework for building enterprise applications.

This article shows how to combine Spring Boot with DeepSeek, a new large language model (LLM) provider, to develop AI-powered services such as chatbots, document processing, or code generation.

Project creation : Using the Spring Initializr, select the Web and AI (OpenAI) starters; the generated pom.xml includes the necessary dependencies.

<properties>
  <java.version>17</java.version>
  <spring-ai.version>1.0.0-M5</spring-ai.version>
  <maven.compiler.source>17</maven.compiler.source>
  <maven.compiler.target>17</maven.compiler.target>
</properties>

<dependencies>
  <dependency>
    <groupId>org.springframework.boot</groupId>
    <artifactId>spring-boot-starter-web</artifactId>
  </dependency>
  <dependency>
    <groupId>org.springframework.ai</groupId>
    <artifactId>spring-ai-openai-spring-boot-starter</artifactId>
  </dependency>
</dependencies>

<dependencyManagement>
  <dependencies>
    <dependency>
      <groupId>org.springframework.ai</groupId>
      <artifactId>spring-ai-bom</artifactId>
      <version>${spring-ai.version}</version>
      <type>pom</type>
      <scope>import</scope>
    </dependency>
  </dependencies>
</dependencyManagement>

Configure OpenAI parameters : Obtain an API key from SiliconFlow (or other providers) and set the key, base URL, and model name in application.yml .

spring:
  ai:
    openai:
      api-key: YOUR_API_KEY
      base-url: https://api.siliconflow.cn
      chat:
        options:
          model: deepseek-ai/DeepSeek-R1-Distill-Llama-8B

The chosen model “deepseek-ai/DeepSeek-R1-Distill-Llama-8B” is the free conversational model provided by SiliconFlow.

Implementing the AI controller : A simple ChatBotController uses ChatClient to forward user messages to the model and return the generated content.

package com.summer.springai.controller;

import groovy.util.logging.Slf4j;
import org.springframework.ai.chat.client.ChatClient;
import org.springframework.web.bind.annotation.*;

@RestController
@CrossOrigin(origins = "*")
@Slf4j
public class ChatBotController {
    private final ChatClient chatClient;

    public ChatBotController(ChatClient.Builder builder) {
        this.chatClient = builder.defaultSystem(
            "You are a weather forecaster. When a date is provided, output Suzhou weather forecast in markdown, ending with: Thank you for your inquiry, I am Yuqing."
        ).build();
    }

    @GetMapping("/chat/{message}")
    public String chat(@PathVariable("message") String message) {
        return chatClient.prompt()
                         .user(message)
                         .call()
                         .content();
    }
}

After building and running the application, accessing http://localhost:8080/ai/chat/2025年2月12日 returns a weather forecast generated by the DeepSeek model.

The article concludes with a brief note that the same approach can be extended to output JSON, store results in a database, or perform data collection for enterprise projects.

Note: The final part of the source contains promotional material unrelated to the technical tutorial.