SpringAI + DeepSeek大模型應用開發

?三、SpringAI

	Spring AI	LangChain4j
Chat	支持	支持
Function	支持	支持
RAG	支持	支持
對話模型	15+	15+
向量模型	10+	15+
向量數據庫	15+	20+
多模態模型	5+	1
JDK	17	8

1. 對話機器人

1.1 快速入門

步驟①：引入依賴（先去掉openai的starter依賴，因為要配置API_KEY）

<?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>3.4.3</version><relativePath/> <!-- lookup parent from repository --></parent><groupId>com.itheima</groupId><artifactId>heima-ai</artifactId><version>0.0.1-SNAPSHOT</version><name>heima-ai</name><description>heima-ai</description><url/><licenses><license/></licenses><developers><developer/></developers><scm><connection/><developerConnection/><tag/><url/></scm><properties><java.version>17</java.version><spring-ai.version>1.0.0-M6</spring-ai.version></properties><dependencies><dependency><groupId>org.springframework.boot</groupId><artifactId>spring-boot-starter-web</artifactId></dependency><dependency><groupId>org.springframework.ai</groupId><artifactId>spring-ai-ollama-spring-boot-starter</artifactId></dependency><dependency><groupId>com.mysql</groupId><artifactId>mysql-connector-j</artifactId><scope>runtime</scope></dependency><dependency><groupId>org.springframework.boot</groupId><artifactId>spring-boot-starter-test</artifactId><scope>test</scope></dependency><dependency><groupId>org.projectlombok</groupId><artifactId>lombok</artifactId><version>1.18.30</version></dependency><dependency><groupId>org.springframework.boot</groupId><artifactId>spring-boot-starter-data-redis</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><build><plugins><plugin><groupId>org.springframework.boot</groupId><artifactId>spring-boot-maven-plugin</artifactId></plugin></plugins></build></project>

注意將lombok的版本改為1.18.30，否則可能出現下面的錯誤

java: java.lang.NoSuchFieldError: Class com.sun.tools.javac.tree.JCTree$JCImport does not have member field 'com.sun.tools.javac.tree.JCTree qualid'

②配置模型

如果用的是ollama

spring:application:name: heima-aiai:ollama:base-url: http://localhost:11434chat:model: deepseek-r1:7b

如果用的是openai

spring:ai:openai:base-url: https://dashscope.aliyuncs.com/compatible-modeapi-key: ${OPENAI_API_KEY}chat:options:model: qwen-max  # 模型名稱temperature: 0.8  # 模型溫度，值越大，輸出結果越隨機

③配置客戶端

package com.itheima.ai.config;import org.springframework.ai.chat.client.ChatClient;
import org.springframework.ai.ollama.OllamaChatModel;
import org.springframework.context.annotation.Bean;
import org.springframework.context.annotation.Configuration;@Configuration
public class CommonConfiguration {@Beanpublic ChatClient chatClient(OllamaChatModel model) {return ChatClient.builder(model).defaultSystem("你是一個熱心、可愛的智能助手，你的名字叫小團團，請以小團團的身份和語氣回答問題。").build();}
}

④發起調用

package com.itheima.ai.controller;import lombok.RequiredArgsConstructor;
import org.springframework.ai.chat.client.ChatClient;
import org.springframework.web.bind.annotation.RequestMapping;
import org.springframework.web.bind.annotation.RestController;@RequiredArgsConstructor
@RestController
@RequestMapping("/ai")
public class ChatController {private final ChatClient chatClient;@RequestMapping("/chat")public String chat(String prompt) {return chatClient.prompt().user(prompt).call().content();}
}

⑤啟動HeimaAiApplication 和 ollama 測試

http://localhost:8080/ai/chat?prompt=%E4%BD%A0%E6%98%AF%E8%B0%81

再次改用流式輸出進行測試：

    @RequestMapping(value = "/chat", produces = "text/html;charset=utf-8")public Flux<String> chat(String prompt) {return chatClient.prompt().user(prompt).stream().content();}

1.2 會話日志

SpringAI利用AOP原理提供了AI會話時的攔截、增強等功能，也就是Advisor。

①修改CommonConfiguration?

package com.itheima.ai.config;import org.springframework.ai.chat.client.ChatClient;
import org.springframework.ai.chat.client.advisor.SimpleLoggerAdvisor;
import org.springframework.ai.ollama.OllamaChatModel;
import org.springframework.context.annotation.Bean;
import org.springframework.context.annotation.Configuration;@Configuration
public class CommonConfiguration {@Beanpublic ChatClient chatClient(OllamaChatModel model) {return ChatClient.builder(model).defaultSystem("你是一個熱心、可愛的智能助手，你的名字叫小團團，請以小團團的身份和語氣回答問題。").defaultAdvisors(new SimpleLoggerAdvisor(),).build();}
}

②在application.yaml添加日志級別

logging:level:org.springframework.ai: debugcom.itheima.ai: debug

2025-06-17T10:18:45.596+08:00 DEBUG 32052 --- [heima-ai] [oundedElastic-1] o.s.a.c.c.advisor.SimpleLoggerAdvisor ? ?: request: AdvisedRequest[chatModel=org.springframework.ai.ollama.OllamaChatModel@614e1bca, userText=你是誰, systemText=你是一個熱心、可愛的智能助手，你的名字叫小團團，請以小團團的身份和語氣回答問題。, chatOptions=org.springframework.ai.ollama.api.OllamaOptions@6913c33b, media=[], functionNames=[], functionCallbacks=[], messages=[], userParams={}, systemParams={}, advisors=[org.springframework.ai.chat.client.DefaultChatClient$DefaultChatClientRequestSpec$1@54c6b5de, org.springframework.ai.chat.client.DefaultChatClient$DefaultChatClientRequestSpec$2@7a8b4de2, SimpleLoggerAdvisor, org.springframework.ai.chat.client.advisor.MessageChatMemoryAdvisor@799e357a, org.springframework.ai.chat.client.DefaultChatClient$DefaultChatClientRequestSpec$1@681ff1c8, org.springframework.ai.chat.client.DefaultChatClient$DefaultChatClientRequestSpec$2@916c150], advisorParams={}, adviseContext={}, toolContext={}]

1.3 會話記憶

步驟①：把資料中的spring-ai-nginx壓縮包放到不含中文和特殊字符的文件夾下并解壓：

②通過cmd運行項目

http://localhost:5173/

③配置CORS策略

package com.itheima.ai.config;import org.springframework.context.annotation.Configuration;
import org.springframework.web.servlet.config.annotation.CorsRegistry;
import org.springframework.web.servlet.config.annotation.WebMvcConfigurer;@Configuration
public class MvcConfiguration implements WebMvcConfigurer {@Overridepublic void addCorsMappings(CorsRegistry registry) {registry.addMapping("/**").allowedOrigins("*").allowedMethods("GET", "POST", "PUT", "DELETE", "OPTIONS").allowedHeaders("*").exposedHeaders("Content-Disposition");}
}

④測試

大模型是不具備記憶能力的，要想讓大模型記住之前聊天的內容，唯一的辦法就是把之前聊天的內容與新的提示詞一起發給大模型。

步驟①：定義會話存儲方式

方式1：使用Spring官方提供的InMemoryChatMemory
方式2：自己實現ChatMemory接口，把會話存儲到Redis、MongoDB等中

（1）方式1：InMemoryChatMemory

②配置會話記憶Advisor

package com.itheima.ai.config;import org.springframework.ai.chat.client.ChatClient;
import org.springframework.ai.chat.client.advisor.MessageChatMemoryAdvisor;
import org.springframework.ai.chat.client.advisor.SimpleLoggerAdvisor;
import org.springframework.ai.chat.memory.ChatMemory;
import org.springframework.ai.chat.memory.InMemoryChatMemory;
import org.springframework.ai.ollama.OllamaChatModel;
import org.springframework.context.annotation.Bean;@Configuration
public class CommonConfiguration {@Beanpublic ChatMemory chatMemory() {return new InMemoryChatMemory();}@Beanpublic ChatClient chatClient(OllamaChatModel model, ChatMemory chatMemory) {return ChatClient.builder(model).defaultSystem("你是一個熱心、可愛的智能助手，你的名字叫小團團，請以小團團的身份和語氣回答問題。").defaultAdvisors(new SimpleLoggerAdvisor(),new MessageChatMemoryAdvisor(chatMemory)  // 會話記憶).build();}
}

③添加會話id

    @RequestMapping(value = "/chat", produces = "text/html;charset=utf-8")public Flux<String> chat(@RequestParam("prompt") String prompt,@RequestParam("chatId") String chatId) {return chatClient.prompt().user(prompt).advisors(a -> a.param(CHAT_MEMORY_CONVERSATION_ID_KEY, chatId)).stream().content();}

（2）方式2：實現ChatMemory接口，保存到Redis中

定義Msg實體類：

package com.itheima.ai.entity.po;import lombok.AllArgsConstructor;
import lombok.Data;
import lombok.NoArgsConstructor;
import org.springframework.ai.chat.messages.*;import java.util.List;
import java.util.Map;@NoArgsConstructor
@AllArgsConstructor
@Data
public class Msg {MessageType messageType;String text;Map<String, Object> metadata;public Msg(Message message) {this.messageType = message.getMessageType();this.text = message.getText();this.metadata = message.getMetadata();}public Message toMessage() {return switch (messageType) {case SYSTEM -> new SystemMessage(text);case USER -> new UserMessage(text, List.of(), metadata);case ASSISTANT -> new AssistantMessage(text, metadata, List.of(), List.of());default -> throw new IllegalArgumentException("Unsupported message type: " + messageType);};}
}

定義ChatMemory實現類：

package com.itheima.ai.repository;import com.fasterxml.jackson.core.JsonProcessingException;
import com.fasterxml.jackson.databind.ObjectMapper;
import com.itheima.ai.entity.po.Msg;
import lombok.RequiredArgsConstructor;
import org.springframework.ai.chat.memory.ChatMemory;
import org.springframework.ai.chat.messages.Message;
import org.springframework.data.redis.core.StringRedisTemplate;
import org.springframework.stereotype.Component;import java.util.List;@RequiredArgsConstructor
@Component
public class RedisChatMemory implements ChatMemory {private final StringRedisTemplate redisTemplate;private final ObjectMapper objectMapper;private final static String PREFIX = "chat:";@Overridepublic void add(String conversationId, List<Message> messages) {if (messages == null || messages.isEmpty()) {return;}List<String> list = messages.stream().map(Msg::new).map(msg -> {try {return objectMapper.writeValueAsString(msg);} catch (JsonProcessingException e) {throw new RuntimeException(e);}}).toList();redisTemplate.opsForList().leftPushAll(PREFIX + conversationId, list);}@Overridepublic List<Message> get(String conversationId, int lastN) {List<String> list = redisTemplate.opsForList().range(PREFIX + conversationId, 0, lastN);if (list == null || list.isEmpty()) {return List.of();}return list.stream().map(s -> {try {return objectMapper.readValue(s, Msg.class);} catch (JsonProcessingException e) {throw new RuntimeException(e);}}).map(Msg::toMessage).toList();}@Overridepublic void clear(String conversationId) {redisTemplate.delete(PREFIX + conversationId);}
}

引入redis依賴：

        <dependency><groupId>org.springframework.boot</groupId><artifactId>spring-boot-starter-data-redis</artifactId></dependency>

配置會話記憶advisor：

package com.itheima.ai.config;import com.fasterxml.jackson.databind.ObjectMapper;
import com.itheima.ai.repository.RedisChatMemory;
import org.springframework.ai.chat.client.ChatClient;
import org.springframework.ai.chat.client.advisor.MessageChatMemoryAdvisor;
import org.springframework.ai.chat.client.advisor.SimpleLoggerAdvisor;
import org.springframework.ai.chat.memory.ChatMemory;
import org.springframework.ai.chat.memory.InMemoryChatMemory;
import org.springframework.ai.ollama.OllamaChatModel;
import org.springframework.beans.factory.annotation.Autowired;
import org.springframework.context.annotation.Bean;
import org.springframework.context.annotation.Configuration;
import org.springframework.data.redis.core.StringRedisTemplate;@Configuration
public class CommonConfiguration {@Autowiredprivate StringRedisTemplate redisTemplate;private ObjectMapper objectMapper = new ObjectMapper();@Beanpublic ChatMemory chatMemory() {return new RedisChatMemory(redisTemplate, objectMapper);}@Beanpublic ChatClient chatClient(OllamaChatModel model, ChatMemory chatMemory) {return ChatClient.builder(model).defaultSystem("你是一個熱心、可愛的智能助手，你的名字叫小團團，請以小團團的身份和語氣回答問題。").defaultAdvisors(new SimpleLoggerAdvisor(),new MessageChatMemoryAdvisor(chatMemory)  // 會話記憶).build();}
}

配置redis連接信息：

spring:data:redis:host: localhost

1.4 會話歷史

接口說明

	查詢會話記錄列表	查詢會話記錄詳情
請求方式	GET	GET
請求路徑	/ai/history/{type}	/ai/history/{type}/{chatId}
請求參數	type: 業務類型	type: 業務類型; chatId: 會話id
返回值	["1234", "1246", "1248"]	[{role: "user", content: ""}]

方式1：保存到JVM內存
方式2：保存到Redis

步驟①：定義會話歷史接口ChatHistoryRepository?

package com.itheima.ai.repository;import java.util.List;public interface ChatHistoryRepository {/*** 保存會話記錄* @param type 業務類型，如：chat、service、pdf* @param chatId 會話ID*/void save(String type, String chatId);/*** 獲取會話ID列表* @param type 業務類型，如：chat、service、pdf* @return 會話ID列表*/List<String> getChatIds(String type);
}

②創建實現類

package com.itheima.ai.repository;import lombok.RequiredArgsConstructor;
import lombok.extern.slf4j.Slf4j;
import org.springframework.ai.chat.memory.ChatMemory;
import org.springframework.stereotype.Component;import java.util.*;@Slf4j
@Component
@RequiredArgsConstructor
public class InMemoryChatHistoryRepository implements ChatHistoryRepository {private Map<String, List<String>> chatHistory;private final ObjectMapper objectMapper;private final ChatMemory chatMemory;@Overridepublic void save(String type, String chatId) {/*if (!chatHistory.containsKey(type)) {chatHistory.put(type, new ArrayList<>());}List<String> chatIds = chatHistory.get(type);*/List<String> chatIds = chatHistory.computeIfAbsent(type, k -> new ArrayList<>());if (chatIds.contains(chatId)) {return;}chatIds.add(0, chatId);}@Overridepublic List<String> getChatIds(String type) {/*List<String> chatIds = chatHistory.get(type);return chatIds == null ? List.of() : chatIds;*/return chatHistory.getOrDefault(type, List.of());}
}

③請求模型前保存會話id

    @RequestMapping(value = "/chat", produces = "text/html;charset=utf-8")public Flux<String> chat(@RequestParam("prompt") String prompt,@RequestParam("chatId") String chatId) {// 1. 保存會話idchatHistoryRepository.save("chat", chatId);// 2. 請求模型return chatClient.prompt().user(prompt).advisors(a -> a.param(CHAT_MEMORY_CONVERSATION_ID_KEY, chatId)).stream().content();}

④定義MessageVo

package com.itheima.ai.entity.vo;import lombok.Data;
import lombok.NoArgsConstructor;
import org.springframework.ai.chat.messages.Message;@NoArgsConstructor
@Data
public class MessageVO {private String role;private String content;public MessageVO(Message message) {this.role = switch (message.getMessageType()) {case USER -> "user";case ASSISTANT -> "assistant";case SYSTEM -> "system";default -> "";};this.content = message.getText();}
}

⑤根據業務類型、會話id查詢會話歷史詳情

package com.itheima.ai.controller;import com.itheima.ai.entity.vo.MessageVO;
import com.itheima.ai.repository.ChatHistoryRepository;
import lombok.RequiredArgsConstructor;
import org.springframework.ai.chat.memory.ChatMemory;
import org.springframework.ai.chat.messages.Message;
import org.springframework.web.bind.annotation.GetMapping;
import org.springframework.web.bind.annotation.PathVariable;
import org.springframework.web.bind.annotation.RequestMapping;
import org.springframework.web.bind.annotation.RestController;import java.util.List;@RequiredArgsConstructor
@RestController
@RequestMapping("/ai/history")
public class ChatHistoryController {private final ChatHistoryRepository chatHistoryRepository;private final ChatMemory chatMemory;/*** 查詢會話歷史列表* @param type 業務類型，如：chat,service,pdf* @return chatId列表*/@GetMapping("/{type}")public List<String> getChatIds(@PathVariable("type") String type) {return chatHistoryRepository.getChatIds(type);}/*** 根據業務類型、chatId查詢會話歷史詳情* @param type 業務類型，如：chat,service,pdf* @param chatId 會話id* @return 指定會話的歷史消息*/@GetMapping("/{type}/{chatId}")public List<MessageVO> getChatHistory(@PathVariable("type") String type, @PathVariable("chatId") String chatId) {List<Message> messages = chatMemory.get(chatId, Integer.MAX_VALUE);if(messages == null) {return List.of();}return messages.stream().map(MessageVO::new).toList();}
}

⑥重啟進行HeimaAiApplication測試

⑦也可以選擇使用Redis來實現

package com.itheima.ai.repository;import lombok.RequiredArgsConstructor;
import org.springframework.data.redis.core.StringRedisTemplate;
import org.springframework.stereotype.Component;import java.util.Collections;
import java.util.List;
import java.util.Set;@RequiredArgsConstructor
@Component
public class RedisChatHistory implements ChatHistoryRepository{private final StringRedisTemplate redisTemplate;private final static String CHAT_HISTORY_KEY_PREFIX = "chat:history:";@Overridepublic void save(String type, String chatId) {redisTemplate.opsForSet().add(CHAT_HISTORY_KEY_PREFIX + type, chatId);}@Overridepublic List<String> getChatIds(String type) {Set<String> chatIds = redisTemplate.opsForSet().members(CHAT_HISTORY_KEY_PREFIX + type);if(chatIds == null || chatIds.isEmpty()) {return Collections.emptyList();}return chatIds.stream().sorted(String::compareTo).toList();}
}