秒懂 Spring AI Alibaba 模型接入流程

本文只讲最基础的模型接入配置依赖 → 配置密钥 → 调用接口三步跑通同步、流式、带记忆、自定义参数四种对话方式。完整代码 Demo 见资源包可自行下载。一、配置 POM 文件1. 指定框架版本号properties java.version17/java.version spring-ai.version1.1.0/spring-ai.version spring-ai-alibaba.version1.1.0.0/spring-ai-alibaba.version spring-ai-alibaba-extensions.version1.1.2.1/spring-ai-alibaba-extensions.version /properties2. 依赖管理引入 BOM通过 BOM 统一管理版本后续引入具体依赖时无需再写版本号避免版本冲突。dependencyManagement dependencies !-- Spring AI 官方 BOM -- dependency groupIdorg.springframework.ai/groupId artifactIdspring-ai-bom/artifactId version${spring-ai.version}/version typepom/type scopeimport/scope /dependency !-- Spring AI Alibaba BOM -- dependency groupIdcom.alibaba.cloud.ai/groupId artifactIdspring-ai-alibaba-bom/artifactId version${spring-ai-alibaba.version}/version typepom/type scopeimport/scope /dependency !-- Spring AI Alibaba 扩展 BOM -- dependency groupIdcom.alibaba.cloud.ai/groupId artifactIdspring-ai-alibaba-extensions-bom/artifactId version${spring-ai-alibaba-extensions.version}/version typepom/type scopeimport/scope /dependency /dependencies /dependencyManagement3. 引入具体依赖dependencies !-- Web 依赖 -- dependency groupIdorg.springframework.boot/groupId artifactIdspring-boot-starter-web/artifactId /dependency !-- Spring AI Alibaba DashScope 依赖 -- dependency groupIdcom.alibaba.cloud.ai/groupId artifactIdspring-ai-alibaba-starter-dashscope/artifactId /dependency !-- 测试依赖 -- dependency groupIdorg.springframework.boot/groupId artifactIdspring-boot-starter-test/artifactId scopetest/scope /dependency /dependencies二、配置 YML 文件server: port: 18080 spring: application: name: spring-ai-alibaba-demo ai: dashscope: api-key: ${AI_DASHSCOPE_API_KEY:} # 从环境变量读取避免硬编码 chat: options: model: qwen-plus # 默认使用 qwen-plus 模型提示api-key建议通过环境变量AI_DASHSCOPE_API_KEY注入不要直接写死在配置文件里。获取密钥大模型服务平台百炼控制台 → 创建 / 查看 API Key。三、调用接口项目启动后服务端口为18080资源包中已包含对应 jar 包可直接访问以下接口功能接口地址同步对话http://localhost:18080/helloworld/simple/chat?query你好流式对话http://localhost:18080/helloworld/stream/chat?query你好带会话记忆http://localhost:18080/helloworld/advisor/chat/123?query我叫小明自定义参数http://localhost:18080/helloworld/advisor/newChat?query你好temperature0.8topP0.7同步对话一次性返回完整结果。流式对话逐字返回体验更接近 ChatGPT。带会话记忆路径中的123为会话 ID同一 ID 可记住上下文。自定义参数可动态调整temperature随机性、topP采样范围等模型参数。