1. 准备工作

1.1. 构建项目，添加pom文件

<?xml version="1.0" encoding="UTF-8"?>
<project xmlns="http://maven.apache.org/POM/4.0.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
         xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 https://maven.apache.org/xsd/maven-4.0.0.xsd">
    <modelVersion>4.0.0</modelVersion>
    <parent>
        <groupId>org.springframework.boot</groupId>
        <artifactId>spring-boot-starter-parent</artifactId>
        <version>3.5.0</version>
        <relativePath/>
    </parent>

    <groupId>cn.cjc</groupId>
    <artifactId>springboot-ai</artifactId>
    <version>1.0-SNAPSHOT</version>

    <properties>
        <maven.compiler.source>17</maven.compiler.source>
        <maven.compiler.target>17</maven.compiler.target>
        <project.build.sourceEncoding>UTF-8</project.build.sourceEncoding>
    </properties>

    <dependencies>
        <dependency>
            <groupId>org.springframework.boot</groupId>
            <artifactId>spring-boot-starter-web</artifactId>
        </dependency>
        <dependency>
            <groupId>org.springframework.boot</groupId>
            <artifactId>spring-boot-starter-test</artifactId>
            <scope>test</scope>
        </dependency>


        <!-- https://mvnrepository.com/artifact/dev.langchain4j/langchain4j -->
        <!--  LangChain4j 核心库 -->
        <dependency>
            <groupId>dev.langchain4j</groupId>
            <artifactId>langchain4j</artifactId>
            <version>1.9.1</version>
        </dependency>
        <!-- AiService依赖 -->
        <dependency>
            <groupId>dev.langchain4j</groupId>
            <artifactId>langchain4j-spring-boot-starter</artifactId>
            <version>1.0.1-beta6</version>
        </dependency>

        <dependency>
            <groupId>dev.langchain4j</groupId>
            <artifactId>langchain4j-core</artifactId>
            <version>1.9.1</version>
        </dependency>
        <!-- LangChain4j OpenAI支持（可用于通义千问的OpenAI兼容接口） -->
        <dependency>
            <groupId>dev.langchain4j</groupId>
            <artifactId>langchain4j-open-ai-spring-boot-starter</artifactId>
            <version>1.9.1-beta17</version>
        </dependency>
        <!--  导入响应式编程依赖包-->
        <dependency>
            <groupId>dev.langchain4j</groupId>
            <artifactId>langchain4j-reactor</artifactId>
            <version>1.9.1-beta17</version>
        </dependency>

        <dependency>
            <groupId>org.projectlombok</groupId>
            <artifactId>lombok</artifactId>
            <version>1.18.22</version>
        </dependency>
    </dependencies>
</project>

1.2. 配置文件

spring:
  application:
    name: springboot-ai
  main:
    allow-bean-definition-overriding: true
    
langchain4j:
  open-ai:
    chat-model:
      api-key: ******
      model-name: qwen-plus
      base-url: https://dashscope.aliyuncs.com/compatible-mode/v1

1.3. 提示词请求类

@Data
public class PromptRequest {
    private String prompt;
    private int userId;
}

2. 高级LLM API的功能定义

• 以下三种能力都集中在名为Assistant的接口中，这些能力由一些注解来辅助实现

• 高级LLM API的好处就是只要做好接口定义，具体实现交给LangChain4j来负责，也就是AiServices.create

public interface MergeChatAssistant {
    
    /**
     * 最简单的对话，只返回助手的回答，不包含任何额外信息
     * @param userMessage 用户消息
     * @return 助手生成的回答
     */
    String simpleChat(String userMessage);

    /**
     * 使用模板进行对话，返回助手的回答
     * @param name 模板中的变量
     * @return 助手生成的回答
     */
    @UserMessage("简单介绍一下{{name}}")
    String templateChat(@V("name") String name);

    /**
     * 设置系统消息，就是设定大模型的身份角色，返回助手的回答
     * @param name 模板中的变量
     * @return 助手生成的回答
     */
    @SystemMessage("你的回答不会超过一百汉字")
    @UserMessage("简单介绍一下{{name}}")
    String templateChatWithSysMsg(@V("name") String name);
}

3. 配置类

• 应用会通过LangChain4jConfig类创建两个实例：openAiChatModel负责低级LLM API，assistant负责高级LLM API

@Data
@Configuration
@ConfigurationProperties(prefix = "langchain4j.open-ai.chat-model")
public class LangChain4jConfig {

    private String apiKey;

    private String modelName;

    private String baseUrl;

    /**
     * 创建并配置OpenAiChatModel实例（使用通义千问的OpenAI兼容接口）
     */
    @Bean
    public OpenAiChatModel openAiChatModel() {
        return OpenAiChatModel.builder()
                .apiKey(apiKey)
                .modelName(modelName)
                .baseUrl(baseUrl)
                .build();
    }

    @Bean
    public MergeChatAssistant assistant(@Qualifier("openAiChatModel") OpenAiChatModel chatModel) {
        return AiServices.create(MergeChatAssistant.class, chatModel);
    }
}

4. 服务类

package cn.cjc.ai.service.impl;

import cn.cjc.ai.service.MergeChatAssistant;
import dev.langchain4j.data.message.AiMessage;
import dev.langchain4j.data.message.ChatMessage;
import dev.langchain4j.data.message.SystemMessage;
import dev.langchain4j.data.message.UserMessage;
import dev.langchain4j.model.chat.request.ChatRequest;
import dev.langchain4j.model.openai.OpenAiChatModel;
import lombok.extern.slf4j.Slf4j;
import org.springframework.beans.factory.annotation.Autowired;
import org.springframework.stereotype.Service;

import java.util.List;


/**
 * 2.集成到Springboot
 */
@Service
@Slf4j
public class MergeChatService {

    @Autowired
    private OpenAiChatModel openAiChatModel;

    /**
     * 获取AI模型的响应（用于接口调用）
     *
     * @param prompt 用户提示词
     * @return AI生成的回答
     */
    public String getResponse(String prompt) {
        return openAiChatModel.chat(prompt);
    }

    @Autowired
    private MergeChatAssistant assistant;

    /**
     * 调用AiService进行最简单的对话
     *
     * @param prompt 用户提示词
     * @return 助手生成的回答
     */
    public String simpleChat(String prompt) {
        return assistant.simpleChat(prompt) + "[from aiservice simpleChat]";
    }

    /**
     * 调用AiService进行模板对话
     *
     * @param name 模板中的变量
     * @return 助手生成的回答
     */
    public String templateChat(String name) {
        return assistant.templateChat(name) + "[from aiservice templateChat]";
    }

    /**
     * 调用AiService进行模板对话，包含系统消息
     *
     * @param name 模板中的变量
     * @return 助手生成的回答
     */
    public String templateChatWithSysMsg(String name) {
        return assistant.templateChatWithSysMsg(name) + "[from aiservice templateChatWithSysMsg]";
    }

    /**
     * 模拟多轮对话
     *
     * @param prompt 模板中的变量
     * @return 助手生成的回答
     */
    public String simulateMultiRoundChat(String prompt) {
        List<ChatMessage> history = List.of(
                SystemMessage.from("你是历史学者，回答问题是简洁风格"),
                UserMessage.from("介绍曹操是谁"),
                AiMessage.from("曹操（155－220年），东汉末年杰出政治家、军事家、文学家，魏国奠基者。他统一北方，推行屯田，唯才是举，善用兵法，亦为建安文学代表人物，著有《观沧海》等诗作。"),
                UserMessage.from(prompt));

        AiMessage reply = openAiChatModel.chat(history).aiMessage();

        return reply.text() + "[from simulateMultiRoundChat]";
    }

    public String useChatRequest(String prompt) {
        List<ChatMessage> messages = List.of(
                SystemMessage.from("你是Java程序员，回答问题是简洁风格"),
                UserMessage.from(prompt));

        ChatRequest request = ChatRequest.builder()
                .messages(messages)
                .temperature(0.7)
                .maxOutputTokens(100)
                .build();

        return openAiChatModel.chat(request).aiMessage().text() + "[from useChatRequest]";
    }
}

5. controller

@RestController
@RequestMapping("/api/mergeChat")
public class MergeChatController {

    @Autowired
    private MergeChatService mergeChatService;

    @PostMapping("/chat")
    public ResponseEntity<String> chat(@RequestBody PromptRequest request) {
        // 检查请求体是否有效
        if (request == null || request.getPrompt() == null || request.getPrompt().trim().isEmpty()) {
            return ResponseEntity.badRequest().body("提示词不能为空");
        }
        try {
            // 调用mergeChatService获取模型响应
            String response = mergeChatService.getResponse(request.getPrompt());
            return ResponseEntity.ok(response);
        } catch (Exception e) {
            // 捕获异常并返回错误信息
            return ResponseEntity.status(500).body("请求处理失败: " + e.getMessage());
        }
    }

    @PostMapping("/simpleChat")
    public ResponseEntity<String> simpleChat(@RequestBody PromptRequest request) {
        String response = mergeChatService.simpleChat(request.getPrompt());
        return ResponseEntity.ok(response);
    }

    @PostMapping("/templateChat")
    public ResponseEntity<String> templateChat(@RequestBody PromptRequest request) {
        String response = mergeChatService.templateChat(request.getPrompt());
        return ResponseEntity.ok(response);
    }

    @PostMapping("/templateChatWithSysMsg")
    public ResponseEntity<String> templateChatWithSysMsg(@RequestBody PromptRequest request) {
        String response = mergeChatService.templateChatWithSysMsg(request.getPrompt());
        return ResponseEntity.ok(response);
    }

    @PostMapping("/simulatemultiroundchat")
    public ResponseEntity<String> simulateMultiRoundChat(@RequestBody PromptRequest request) {
        String response = mergeChatService.simulateMultiRoundChat(request.getPrompt());
        return ResponseEntity.ok(response);
    }

    @PostMapping("/usechatrequest")
    public ResponseEntity<String> useChatRequest(@RequestBody PromptRequest request) {
        String response = mergeChatService.useChatRequest(request.getPrompt());
        return ResponseEntity.ok(response);
    }
}