SpringBoot 接入 豆包 火山方舟大模型

火山方舟控制台 开通模型推理、知识库 应用入口；

文档中心 各类接口说明及SDK 获取；

向量数据库VikingDB 文档 下翻找到有java操作案例；

实现目标功能效果：

1. 通过SDK调用 豆包大模型，在代码内实现问答的效果；
   官方示例

2. 通过使用知识库，自己上传文档。让豆包大模型可以回答 你文档中的内容数据（即自定义属于你自己（自身业务领域）的大模型）；

前置步骤：

• 点击右上角头像处去创建 API 访问密钥；
• 控制台内 模型推理 - 在线推理 创建一个推理接入点（按提示操作即可）；
• 控制台内 知识库 - 创建知识库；

代码

package com.example.demo.service.doubao;import com.volcengine.ark.runtime.model.completion.chat.ChatCompletionRequest;
import com.volcengine.ark.runtime.model.completion.chat.ChatMessage;
import com.volcengine.ark.runtime.model.completion.chat.ChatMessageRole;
import com.volcengine.ark.runtime.service.ArkService;
import com.volcengine.auth.ISignerV4;
import com.volcengine.auth.impl.SignerV4Impl;
import com.volcengine.model.Credentials;
import com.volcengine.service.SignableRequest;
import org.apache.http.HttpResponse;
import org.apache.http.NameValuePair;
import org.apache.http.client.HttpClient;
import org.apache.http.client.config.RequestConfig;
import org.apache.http.client.utils.URIBuilder;
import org.apache.http.entity.StringEntity;
import org.apache.http.impl.client.HttpClients;
import org.apache.http.util.EntityUtils;import java.util.ArrayList;
import java.util.List;/*** @Author：wwf* @Package：com.example.demo.service* @Project：demo* @name：CreateArkClientExample* @Date：2024/11/28 15:32* @Filename：CreateArkClientExample*/
public class CreateArkClientExample {//模型推理 - 在线推理 - API 调用 内查看获取private static final String API_KEY = "ea9fxxxxxxxxx49c";//这里获取 https://console.volcengine.com/iam/keymanageprivate static final String AK = "AKLTZDI2Yxxxxxxxxx2U2NGZmMzA";private static final String SK = "Tnpjd0xxxxxxxxxOa1pXRQ";public static void main(String[] args) throws Exception {// 以华北 2 (北京) 为例，<ARK_DOMAIN> 处应改为 ark.cn-beijing.volces.comArkService service = ArkService.builder().apiKey(API_KEY).baseUrl("https://ark.cn-beijing.volces.com/api/v3").build();System.out.println("
----- standard request -----");String host = "api-knowledgebase.mlp.cn-beijing.volces.com";String path = "/api/knowledge/collection/search_knowledge";String method = "POST";String body = "{"name":"wwfTest","query":"总结介绍下王清"}";SignableRequest request = prepareRequest(host, path, method, null, body, AK, SK);System.out.println(request.getURI());HttpClient client = HttpClients.createDefault();HttpResponse response = client.execute(request);int statusCode = response.getStatusLine().getStatusCode();String content = EntityUtils.toString(response.getEntity());System.out.println(statusCode);System.out.println(content);final List<ChatMessage> streamMessages = new ArrayList<>();final ChatMessage streamSystemMessage = ChatMessage.builder().role(ChatMessageRole.SYSTEM).content("你是豆包，是由字节跳动开发的 AI 人工智能助手").content(content).build();final ChatMessage streamUserMessage = ChatMessage.builder().role(ChatMessageRole.USER).content("总结介绍下王清").build();streamMessages.add(streamSystemMessage);streamMessages.add(streamUserMessage);ChatCompletionRequest streamChatCompletionRequest = ChatCompletionRequest.builder().model("ep-20241128155250-zlbd8")//回调函数
//                .tools().messages(streamMessages).build();service.streamChatCompletion(streamChatCompletionRequest).doOnError(Throwable::printStackTrace).blockingForEach(choice -> {if (choice.getChoices().size() > 0) {System.out.print(choice.getChoices().get(0).getMessage().getContent());}});// shutdown serviceservice.shutdownExecutor();}/*** 构建请求** @param host* @param path* @param method* @param params* @param body* @param ak* @param sk* @return* @throws Exception*/public static SignableRequest prepareRequest(String host, String path, String method, List<NameValuePair> params, String body, String ak, String sk) throws Exception {SignableRequest request = new SignableRequest();request.setMethod(method);request.setHeader("Accept", "application/json");request.setHeader("Content-Type", "application/json");request.setHeader("Host", "api-knowledgebase.mlp.cn-beijing.volces.com");request.setEntity(new StringEntity(body, "utf-8"));URIBuilder builder = request.getUriBuilder();builder.setScheme("https");builder.setHost(host);builder.setPath(path);if (params != null) {builder.setParameters(params);}RequestConfig requestConfig = RequestConfig.custom().setSocketTimeout(5000).setConnectTimeout(2000).build();request.setConfig(requestConfig);Credentials credentials = new Credentials("cn-north-1", "air");credentials.setAccessKeyID(ak);credentials.setSecretAccessKey(sk);// 签名ISignerV4 ISigner = new SignerV4Impl();ISigner.sign(request, credentials);return request;}
}

示例代码说明：

本人随意使用gpt 编写了一篇短片小说，通过平台内知识库，以文档形式上传进去。小说内有个人物叫 王清，于是问 豆包模型 介绍下王清的信息，此时 模型可以准确快速的响应 关于王清的信息；

，，，，，，，，，
本文为个人操作接入记录，官方文档示例还是比较全的。仅记录大致接入流程；