Build Your First AI Chatbot with Spring Boot and DeepSeek LLM

Why Combine Spring Boot and DeepSeek?

Artificial intelligence is reshaping software development, and Spring Boot offers a fast, convention‑over‑configuration way to build backend services. DeepSeek provides a cost‑effective large language model (LLM) that rivals many industry leaders. Integrating the two lets developers create intelligent applications such as chatbots, document processors, or code generators with minimal boilerplate.

Step 1: Create a Spring Boot Project

Using the Spring Initializr, start a new project and select the Web and OpenAI starters. After generation, the project automatically includes the necessary dependencies.

Step 2: Add Required Maven 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>

Step 3: Configure DeepSeek (via SiliconFlow)

Register at the SiliconFlow platform (e.g., https://cloud.siliconflow.cn/i/pCa1dBVX) and generate an API key. The key and the chosen model are then placed 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 model name deepseek-ai/DeepSeek-R1-Distill-Llama-8B identifies the specific DeepSeek LLM you will call.

Step 4: Implement the Chat Controller

The following Java class creates a ChatClient with a system prompt that makes the model act as a weather forecaster for Suzhou. The REST endpoint /chat/{message} forwards the user’s date string to the model and returns 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(
            "你是一个天气预报员，当有人输入日期的时候，你输出苏州的天气预报信息，"
            + "生成结果在html页面中以markdown的格式输出，最后输出结尾的时候始终以下面的语句结尾：感谢您的咨询，我是舆情君。"
        ).build();
    }

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

Step 5: Run and Test

Start the Spring Boot application and request the endpoint, for example: http://localhost:8080/ai/chat/2025年2月12日 The response returns a markdown‑formatted weather forecast for the given date, ending with the fixed sign‑off “感谢您的咨询，我是舆情君”。

This simple demonstration shows how a Spring Boot service can leverage DeepSeek’s LLM to provide domain‑specific AI functionality. In a real project you could extend the controller to return JSON, store results in a database, or chain multiple model calls for more complex workflows.