【功能新增】AI：集成 Azure 的 OpenAI 模型

YunaiV
1 parent b4538568
Showing 7 changed files with 47 additions and 6 deletions
yudao-module-ai/yudao-module-ai-biz/src/main/java/cn/iocoder/yudao/module/ai/service/knowledge/DocServiceImpl.java
yudao-module-ai/yudao-spring-boot-starter-ai/pom.xml
yudao-module-ai/yudao-spring-boot-starter-ai/src/main/java/cn/iocoder/yudao/framework/ai/config/YudaoAiAutoConfiguration.java
yudao-module-ai/yudao-spring-boot-starter-ai/src/main/java/cn/iocoder/yudao/framework/ai/core/enums/AiPlatformEnum.java
yudao-module-ai/yudao-spring-boot-starter-ai/src/main/java/cn/iocoder/yudao/framework/ai/core/factory/AiModelFactoryImpl.java
yudao-module-ai/yudao-spring-boot-starter-ai/src/main/java/cn/iocoder/yudao/framework/ai/core/util/AiUtils.java
yudao-server/src/main/resources/application.yaml
@@ -7,7 +7,6 @@ import org.springframework.ai.reader.tika.TikaDocumentReader;
 import org.springframework.ai.transformer.splitter.TokenTextSplitter;
 import org.springframework.ai.vectorstore.RedisVectorStore;
 import org.springframework.beans.factory.annotation.Value;
-import org.springframework.stereotype.Service;
  
 import java.util.List;
  
@@ -16,7 +15,7 @@ import java.util.List;
  *
  * @author xiaoxin
  */
-@Service
+//@Service  // TODO 芋艿：临时注释，避免无法启动
 @Slf4j
 public class DocServiceImpl implements DocService {
  
@@ -23,7 +23,6 @@
             <artifactId>spring-ai-zhipuai-spring-boot-starter</artifactId>
             <version>${spring-ai.version}</version>
         </dependency>
-
         <dependency>
             <groupId>org.springframework.ai</groupId>
             <artifactId>spring-ai-openai-spring-boot-starter</artifactId>
@@ -31,6 +30,11 @@
         </dependency>
         <dependency>
             <groupId>org.springframework.ai</groupId>
+            <artifactId>spring-ai-azure-openai-spring-boot-starter</artifactId>
+            <version>${spring-ai.version}</version>
+        </dependency>
+        <dependency>
+            <groupId>org.springframework.ai</groupId>
             <artifactId>spring-ai-ollama-spring-boot-starter</artifactId>
             <version>${spring-ai.version}</version>
         </dependency>
@@ -12,6 +12,7 @@ import com.alibaba.cloud.ai.tongyi.TongYiAutoConfiguration;
 import lombok.extern.slf4j.Slf4j;
 import org.springframework.ai.autoconfigure.vectorstore.redis.RedisVectorStoreProperties;
 import org.springframework.ai.document.MetadataMode;
+import org.springframework.ai.embedding.EmbeddingModel;
 import org.springframework.ai.transformer.splitter.TokenTextSplitter;
 import org.springframework.ai.transformers.TransformersEmbeddingModel;
 import org.springframework.ai.vectorstore.RedisVectorStore;
@@ -21,6 +22,7 @@ import org.springframework.boot.autoconfigure.data.redis.RedisProperties;
 import org.springframework.boot.context.properties.EnableConfigurationProperties;
 import org.springframework.context.annotation.Bean;
 import org.springframework.context.annotation.Import;
+import org.springframework.context.annotation.Lazy;
 import redis.clients.jedis.JedisPooled;
  
 /**
@@ -82,7 +84,8 @@ public class YudaoAiAutoConfiguration {
  
     // ========== rag 相关 ==========
     @Bean
-    public TransformersEmbeddingModel transformersEmbeddingClient() {
+    @Lazy // TODO 芋艿：临时注释，避免无法启动
+    public EmbeddingModel transformersEmbeddingClient() {
         return new TransformersEmbeddingModel(MetadataMode.EMBED);
     }
  
@@ -90,6 +93,7 @@ public class YudaoAiAutoConfiguration {
      * 我们启动有加载很多 Embedding 模型，不晓得取哪个好，先 new 个 TransformersEmbeddingModel 跑
      */
     @Bean
+    @Lazy // TODO 芋艿：临时注释，避免无法启动
     public RedisVectorStore vectorStore(TransformersEmbeddingModel transformersEmbeddingModel, RedisVectorStoreProperties properties,
                                         RedisProperties redisProperties) {
         var config = RedisVectorStore.RedisVectorStoreConfig.builder()
@@ -105,6 +109,7 @@ public class YudaoAiAutoConfiguration {
     }
  
     @Bean
+    @Lazy // TODO 芋艿：临时注释，避免无法启动
     public TokenTextSplitter tokenTextSplitter() {
         return new TokenTextSplitter(500, 100, 5, 10000, true);
     }
@@ -22,7 +22,8 @@ public enum AiPlatformEnum {
  
     // ========== 国外平台 ==========
  
-    OPENAI("OpenAI", "OpenAI"),
+    OPENAI("OpenAI", "OpenAI"), // OpenAI 官方
+    AZURE_OPENAI("AzureOpenAI", "AzureOpenAI"), // OpenAI 微软
     OLLAMA("Ollama", "Ollama"),
  
     STABLE_DIFFUSION("StableDiffusion", "StableDiffusion"), // Stability AI
@@ -21,6 +21,10 @@ import com.alibaba.cloud.ai.tongyi.image.TongYiImagesModel;
 import com.alibaba.cloud.ai.tongyi.image.TongYiImagesProperties;
 import com.alibaba.dashscope.aigc.generation.Generation;
 import com.alibaba.dashscope.aigc.imagesynthesis.ImageSynthesis;
+import com.azure.ai.openai.OpenAIClient;
+import org.springframework.ai.autoconfigure.azure.openai.AzureOpenAiAutoConfiguration;
+import org.springframework.ai.autoconfigure.azure.openai.AzureOpenAiChatProperties;
+import org.springframework.ai.autoconfigure.azure.openai.AzureOpenAiConnectionProperties;
 import org.springframework.ai.autoconfigure.ollama.OllamaAutoConfiguration;
 import org.springframework.ai.autoconfigure.openai.OpenAiAutoConfiguration;
 import org.springframework.ai.autoconfigure.qianfan.QianFanAutoConfiguration;
@@ -31,6 +35,7 @@ import org.springframework.ai.autoconfigure.zhipuai.ZhiPuAiAutoConfiguration;
 import org.springframework.ai.autoconfigure.zhipuai.ZhiPuAiChatProperties;
 import org.springframework.ai.autoconfigure.zhipuai.ZhiPuAiConnectionProperties;
 import org.springframework.ai.autoconfigure.zhipuai.ZhiPuAiImageProperties;
+import org.springframework.ai.azure.openai.AzureOpenAiChatModel;
 import org.springframework.ai.chat.model.ChatModel;
 import org.springframework.ai.image.ImageModel;
 import org.springframework.ai.model.function.FunctionCallbackContext;
@@ -82,6 +87,8 @@ public class AiModelFactoryImpl implements AiModelFactory {
                     return buildXingHuoChatModel(apiKey);
                 case OPENAI:
                     return buildOpenAiChatModel(apiKey, url);
+                case AZURE_OPENAI:
+                    return buildAzureOpenAiChatModel(apiKey, url);
                 case OLLAMA:
                     return buildOllamaChatModel(url);
                 default:
@@ -106,6 +113,8 @@ public class AiModelFactoryImpl implements AiModelFactory {
                 return SpringUtil.getBean(XingHuoChatModel.class);
             case OPENAI:
                 return SpringUtil.getBean(OpenAiChatModel.class);
+            case AZURE_OPENAI:
+                return SpringUtil.getBean(AzureOpenAiChatModel.class);
             case OLLAMA:
                 return SpringUtil.getBean(OllamaChatModel.class);
             default:
@@ -269,6 +278,21 @@ public class AiModelFactoryImpl implements AiModelFactory {
     }
  
     /**
+     * 可参考 {@link AzureOpenAiAutoConfiguration}
+     */
+    private static AzureOpenAiChatModel buildAzureOpenAiChatModel(String apiKey, String url) {
+        AzureOpenAiAutoConfiguration azureOpenAiAutoConfiguration = new AzureOpenAiAutoConfiguration();
+        // 创建 OpenAIClient 对象
+        AzureOpenAiConnectionProperties connectionProperties = new AzureOpenAiConnectionProperties();
+        connectionProperties.setApiKey(apiKey);
+        connectionProperties.setEndpoint(url);
+        OpenAIClient openAIClient = azureOpenAiAutoConfiguration.openAIClient(connectionProperties);
+        // 获取 AzureOpenAiChatProperties 对象
+        AzureOpenAiChatProperties chatProperties = SpringUtil.getBean(AzureOpenAiChatProperties.class);
+        return azureOpenAiAutoConfiguration.azureOpenAiChatModel(openAIClient, chatProperties, null, null);
+    }
+
+    /**
      * 可参考 {@link OpenAiAutoConfiguration}
      */
     private OpenAiImageModel buildOpenAiImageModel(String openAiToken, String url) {
@@ -5,6 +5,7 @@ import cn.iocoder.yudao.framework.ai.core.enums.AiPlatformEnum;
 import cn.iocoder.yudao.framework.ai.core.model.deepseek.DeepSeekChatOptions;
 import cn.iocoder.yudao.framework.ai.core.model.xinghuo.XingHuoChatOptions;
 import com.alibaba.cloud.ai.tongyi.chat.TongYiChatOptions;
+import org.springframework.ai.azure.openai.AzureOpenAiChatOptions;
 import org.springframework.ai.chat.messages.*;
 import org.springframework.ai.chat.prompt.ChatOptions;
 import org.springframework.ai.ollama.api.OllamaOptions;
@@ -35,6 +36,9 @@ public class AiUtils {
                 return XingHuoChatOptions.builder().model(model).temperature(temperatureF).maxTokens(maxTokens).build();
             case OPENAI:
                 return OpenAiChatOptions.builder().withModel(model).withTemperature(temperatureF).withMaxTokens(maxTokens).build();
+            case AZURE_OPENAI:
+                // TODO 芋艿：貌似没 model 字段？？？！
+                return AzureOpenAiChatOptions.builder().withDeploymentName(model).withTemperature(temperatureF).withMaxTokens(maxTokens).build();
             case OLLAMA:
                 return OllamaOptions.create().withModel(model).withTemperature(temperatureF).withNumPredict(maxTokens);
             default:
@@ -162,9 +162,13 @@ spring:
       secret-key: R9mYF9dl9KASgi5RUq0FQt3wRisSnOcK
     zhipuai: # 智谱 AI
       api-key: 32f84543e54eee31f8d56b2bd6020573.3vh9idLJZ2ZhxDEs
-    openai:
+    openai: # OpenAI 官方
       api-key: sk-yzKea6d8e8212c3bdd99f9f44ced1cae37c097e5aa3BTS7z
       base-url: https://api.gptsapi.net
+    azure: # OpenAI 微软
+      openai:
+        endpoint: https://eastusprejade.openai.azure.com
+        api-key: xxx
     ollama:
       base-url: http://127.0.0.1:11434
       chat: