AI2.0 【多模态】 20260615

 

统一封装任意二进制媒体数据，解决不同文件类型、本地 / 网络资源适配问题，双核心属性：

1. mimeType：媒体类型标识，标准 MIME 格式
   • 图像：image/png / image/jpeg
   • 音频：audio/wav / audio/mp3
   • 视频：video/mp4（取决于模型支持）
2. resource：Spring 标准Resource资源对象，兼容 Spring 全套资源体系：
   • ClassPathResource：classpath 内静态图片
   • FileSystemResource：服务器本地磁盘文件
   • UrlResource：网络图片 URL
   • ByteArrayResource：内存二进制字节流（前端上传图片）

        3. ChatClient 流畅式调用 API

Spring AI 2.0 主推的业务层工具，基于建造者模式封装多模态请求全流程，优势：

• 链式调用，无需手动组装UserMessage/Prompt底层对象
• 内置系统提示词、参数配置、工具调用、结构化输出能力
• 自动适配多模态参数传递逻辑，降低多模态代码复杂度

 

 

 

 

 

1、pom

<properties>
    <java.version>17</java.version>
    <spring-ai.version>2.0.0-M5</spring-ai.version>
</properties>
<dependencies>
    <dependency>
        <groupId>org.springframework.boot</groupId>
        <artifactId>spring-boot-starter-webmvc</artifactId>
    </dependency>
    <dependency>
        <groupId>org.springframework.ai</groupId>
        <artifactId>spring-ai-starter-model-openai</artifactId>
    </dependency>
    <dependency>
        <groupId>org.springframework.ai</groupId>
        <artifactId>spring-ai-starter-model-ollama</artifactId>
    </dependency>
    <!-- Redis 向量库-->
    <dependency>
        <groupId>org.springframework.ai</groupId>
        <artifactId>spring-ai-starter-vector-store-redis</artifactId>
    </dependency>
    <!--Redis 客户端-->
    <dependency>
        <groupId>org.springframework.boot</groupId>
        <artifactId>spring-boot-starter-data-redis</artifactId>
        <exclusions>
            <exclusion>
                <groupId>io.lettuce</groupId>
                <artifactId>lettuce-core</artifactId>
            </exclusion>
        </exclusions>
    </dependency>
    <dependency>
        <groupId>redis.clients</groupId>
        <artifactId>jedis</artifactId>
    </dependency>
    <!-- Tika 通用文档读取器 -->
    <dependency>
        <groupId>org.springframework.ai</groupId>
        <artifactId>spring-ai-tika-document-reader</artifactId>
    </dependency>
    <!-- Markdown 文档读取器 -->
    <dependency>
        <groupId>org.springframework.ai</groupId>
        <artifactId>spring-ai-markdown-document-reader</artifactId>
    </dependency>
    <!-- 问答顾问器 advisor-->
    <dependency>
        <groupId>org.springframework.ai</groupId>
        <artifactId>spring-ai-advisors-vector-store</artifactId>
    </dependency>
    <!-- MCP 客户端：通过 SSE 连接本机 mcp-server（默认端口 18089） -->
    <dependency>
        <groupId>org.springframework.ai</groupId>
        <artifactId>spring-ai-starter-mcp-client</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>

 

2、application.yml

2.1

${OPENAI_API_KEY}

 

 

spring:
  autoconfigure:
    exclude:  #禁用 Ollama Embedding（如果你用 OpenAI），不然会冲突
      - org.springframework.ai.model.ollama.autoconfigure.OllamaEmbeddingAutoConfiguration
  application:
    name: dashscope18088
  data:
    redis:
      url: redis://192.168.91.165:6379
      client-type: jedis
  ai:
    openai:
      api-key: ${OPENAI_API_KEY}
      base-url: https://dashscope.aliyuncs.com/compatible-mode/v1
      chat:
        options:
          model: qwen3.6-plus
      embedding:
        options:
          model: text-embedding-v4
          dimensions: 2048 # 输出向量维度2048
        #      chat:
#        model: qwen3.6-plus
    mcp:
      client:
        enabled: true
        type: SYNC
        # 与 mcp-server 一致；须先启动 mcp-server（默认 http://localhost:18089）
        sse:
          connections:
            mcp-weather-server:
              url: http://localhost:18089
    vectorstore:
      redis:
        initialize-schema: true   #是否在启动时创建/初始化向量索引等结构
        index-name: sf2026_ai_vector_index  # RediSearch 里的索引名
        prefix: sf2026   #存向量文档用的Redis键前级
    ollama:
      base-url: http://192.168.91.164:11434
      chat:
        model: qwen2.5:0.5b

#        options:
#          temperature: 0.7
#          max-tokens: 2000
server:
  port: 18088
logging:
  level:
    org.springframework.ai: debug
    org.springframework.ai.mcp: debug

3、controller

 

 

import org.springframework.ai.chat.client.ChatClient;
import org.springframework.beans.factory.annotation.Autowired;
import org.springframework.core.io.ClassPathResource;
import org.springframework.http.MediaType;
import org.springframework.web.bind.annotation.GetMapping;
import org.springframework.web.bind.annotation.RestController;

@RestController
public class MyMultimodalityController {

    @Autowired
    private ChatClient chatClient;

    @GetMapping("/aiimg")
    public String analyzeImage(){
        String result= chatClient.prompt()
                .user(u -> u
                        .text("请详细描述这张图片的内容，包括物体、颜色、场景和可能的用途")
                        .media(MediaType.IMAGE_PNG, new ClassPathResource("apple.png"))
                )
                .call()
                .content();
        System.out.println("【aiimg】"+result);
        return "【aiimg】"+result;
    }
}

 

4、http://localhost:18088/aiimg