一、什么是 Spring AI Alibaba

Spring AI Alibaba 是一款以 Spring AI 为基础，深度集成百炼平台，支持 ChatBot、工作流、多智能体应用开发模式的 AI 框架。

在 1.0 版本中，Spring AI Alibaba 提供以下核心能力，让开发者可以快速构建自己的 Agent、Workflow 或 Multi-Agent 应用。

Graph 多智能体框架。 基于 Spring AI Alibaba Graph，开发者可快速构建工作流、多智能体应用，无需关心流程编排、上下文记忆管理等底层实现。通过 Graph 与低代码、自规划智能体结合，为开发者提供从低代码、高代码到零代码构建智能体的更灵活选择。

通过 AI 生态集成，解决企业智能体落地过程中关心的痛点问题。 Spring AI Alibaba 支持与百炼平台深度集成，提供模型接入、RAG 知识库解决方案；支持 ARMS、Langfuse 等可观测产品无缝接入；支持企业级的 MCP 集成，包括 Nacos MCP Registry 分布式注册与发现、自动 Router 路由等。

探索具备自主规划能力的通用智能体产品与平台。 社区发布了基于 Spring AI Alibaba 框架实现的 JManus 智能体，除了对标 Manus 的通用智能体能力外，我们的目标是基于 JManus 探索自主规划在智能体开发方向的应用，为开发者提供从低代码、高代码到零代码构建智能体的更灵活选择。

官网：https://java2ai.com/
github：https://github.com/alibaba/spring-ai-alibaba

二、Quickly Run a ChatBot

参考：https://java2ai.com/docs/frameworks/studio/quick-start

There’s a ChatBot example provided by the community at examples/chatbot.

1，Download the code.

git clone --depth=1 https://github.com/alibaba/spring-ai-alibaba.git
cd spring-ai-alibaba/examples/chatbot

2，Start the ChatBot.

Before starting, set API-KEY first (visit Aliyun Bailian to get API-KEY):

# this example uses 'spring-ai-alibaba-starter-dashscope', visit https://java2ai.com to learn how to use OpenAI/DeepSeek.
export AI_DASHSCOPE_API_KEY=your-api-key

这里我们修改application.properties使用本地Ollama模型：

# Spring AI Configuration - DashScope (Disabled)
spring.autoconfigure.exclude=\
  com.alibaba.cloud.ai.autoconfigure.dashscope.DashScopeChatAutoConfiguration,\
  com.alibaba.cloud.ai.autoconfigure.dashscope.DashScopeAgentAutoConfiguration,\
  com.alibaba.cloud.ai.autoconfigure.dashscope.DashScopeAudioSpeechAutoConfiguration,\
  com.alibaba.cloud.ai.autoconfigure.dashscope.DashScopeAudioTranscriptionAutoConfiguration

# Spring AI Configuration - Ollama (Enabled)
spring.ai.ollama.base-url=http://localhost:11434
spring.ai.ollama.chat.enabled=true
spring.ai.ollama.chat.model=qwen2.5:0.5b
spring.ai.ollama.chat.temperature=0.7
spring.ai.model.chat=ollama

mvn spring-boot:run

3，Chat with ChatBot.

Open the browser and visit http://localhost:8080/chatui/index.html to chat with the ChatBot.

三、DeepResearch

DeepResearch 是一款基于阿里巴巴 Spring AI 构建的智能体，能够进行深入的研究并撰写高质量的报告。使用 LLM 循环调用工具是最简单的智能体形式，但这可能会导致智能体功能“浅薄”。DeepResearch 实现了任务规划、子智能体、文件系统访问以及详细的提示，从而能够解决复杂的多步骤研究任务。

DeepResearch实现了一种超越简单工具调用循环的智能体架构。它结合了规划、上下文管理和子智能体协作，以应对需要多步骤和深入调查的复杂研究任务。

参考： https://java2ai.com/agents/deepresearch/quick-start

git clone https://github.com/alibaba/spring-ai-alibaba.git
cd spring-ai-alibaba/examples/deepresearch

使用ollama做如下修改，将 AI 模型（ChatModel）的创建方式从 内部硬编码 改为 外部依赖注入，并更新了配置文件：

diff --git a/examples/deepresearch/src/main/java/com/alibaba/cloud/ai/examples/deepresearch/AgentStaticLoader.java b/examples/deepresearch/src/main/java/com/alibaba/cloud/ai/examples/deepresearch/AgentStaticLoader.java
index f99d75c08..471c65c5f 100644
--- a/examples/deepresearch/src/main/java/com/alibaba/cloud/ai/examples/deepresearch/AgentStaticLoader.java
+++ b/examples/deepresearch/src/main/java/com/alibaba/cloud/ai/examples/deepresearch/AgentStaticLoader.java
@@ -21,6 +21,7 @@ import com.alibaba.cloud.ai.graph.GraphRepresentation;
 import com.alibaba.cloud.ai.graph.agent.Agent;
 import com.alibaba.cloud.ai.graph.agent.ReactAgent;

+import org.springframework.ai.chat.model.ChatModel;
 import org.springframework.ai.tool.ToolCallback;
 import org.springframework.ai.tool.ToolCallbackProvider;

@@ -52,13 +53,13 @@ class AgentStaticLoader implements AgentLoader {

 //     public AgentStaticLoader(){}

-       public AgentStaticLoader(ToolCallbackProvider toolCallbackProvider) {
+       public AgentStaticLoader(ChatModel chatModel, ToolCallbackProvider toolCallbackProvider) {

                List<ToolCallback> toolCallbacks = Arrays.asList(toolCallbackProvider.getToolCallbacks());

                System.out.println("Loaded MCP tool callbacks: " + toolCallbacks.size());

-               ReactAgent researchAgent = new DeepResearchAgent().getResearchAgent(toolCallbacks);
+               ReactAgent researchAgent = new DeepResearchAgent(chatModel).getResearchAgent(toolCallbacks);
                GraphRepresentation representation = researchAgent.getAndCompileGraph().stateGraph.getGraph(GraphRepresentation.Type.PLANTUML);

                System.out.println(representation.content());
diff --git a/examples/deepresearch/src/main/java/com/alibaba/cloud/ai/examples/deepresearch/DeepResearchAgent.java b/examples/deepresearch/src/main/java/com/alibaba/cloud/ai/examples/deepresearch/DeepResearchAgent.java
index 77ff317ef..a1a592c34 100644
--- a/examples/deepresearch/src/main/java/com/alibaba/cloud/ai/examples/deepresearch/DeepResearchAgent.java
+++ b/examples/deepresearch/src/main/java/com/alibaba/cloud/ai/examples/deepresearch/DeepResearchAgent.java
@@ -15,8 +15,6 @@
  */
 package com.alibaba.cloud.ai.examples.deepresearch;

-import com.alibaba.cloud.ai.dashscope.api.DashScopeApi;
-import com.alibaba.cloud.ai.dashscope.chat.DashScopeChatModel;
 import com.alibaba.cloud.ai.graph.agent.ReactAgent;
 import com.alibaba.cloud.ai.graph.agent.extension.interceptor.FilesystemInterceptor;
 import com.alibaba.cloud.ai.graph.agent.extension.interceptor.LargeResultEvictionInterceptor;
@@ -66,11 +64,9 @@ public class DeepResearchAgent {

 //     private ShellTool shellTool = ShellTool.builder().build();

-       public DeepResearchAgent() {
-               // Create DashScopeApi instance using the API key from environment variable
-               DashScopeApi dashScopeApi = DashScopeApi.builder().apiKey(System.getenv("AI_DASHSCOPE_API_KEY")).build();
-               // Create DashScope ChatModel instance
-               this.chatModel = DashScopeChatModel.builder().dashScopeApi(dashScopeApi).build();
+       public DeepResearchAgent(ChatModel chatModel) {
+               // Use injected ChatModel (can be DashScope, Ollama, or any other configured model)
+               this.chatModel = chatModel;

                this.systemPrompt = researchInstructions + "\n\n" + BASE_AGENT_PROMPT;

diff --git a/examples/deepresearch/src/main/resources/application.yml b/examples/deepresearch/src/main/resources/application.yml
index 97212ef3b..ebc32088b 100755
--- a/examples/deepresearch/src/main/resources/application.yml
+++ b/examples/deepresearch/src/main/resources/application.yml
@@ -15,6 +15,24 @@ spring:
               # 官方地址：https://mcp.jina.ai 本地部署：http://127.0.0.1:8787
               url: "https://mcp.jina.ai"
               endpoint: "/sse"
+              headers:
+                x-api-key: ${JINA_API_KEY:}
         request-timeout: 60000
+    # Spring AI Configuration - Ollama (Enabled)
+    ollama:
+      base-url: http://localhost:11434
+      chat:
+        enabled: true
+        model: qwen2.5:0.5b
+        temperature: 0.7
+    model:
+      chat: ollama
+  autoconfigure:
+    exclude:
+      - com.alibaba.cloud.ai.autoconfigure.dashscope.DashScopeChatAutoConfiguration
+      - com.alibaba.cloud.ai.autoconfigure.dashscope.DashScopeAgentAutoConfiguration
+      - com.alibaba.cloud.ai.autoconfigure.dashscope.DashScopeAudioSpeechAutoConfiguration
+      - com.alibaba.cloud.ai.autoconfigure.dashscope.DashScopeAudioTranscriptionAutoConfiguration
+