LangChain4j 提供了一系列官方封装的 ONNX 模型包，全部可离线使用。

1 依赖

<?xml version="1.0" encoding="UTF-8"?>
<project xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns="http://maven.apache.org/POM/4.0.0"
         xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 https://maven.apache.org/xsd/maven-4.0.0.xsd">

    <modelVersion>4.0.0</modelVersion>

    <parent>
        <groupId>org.springframework.boot</groupId>
        <artifactId>spring-boot-starter-parent</artifactId>
        <version>3.5.6</version>
        <relativePath/> <!-- lookup parent from repository -->
    </parent>

    <groupId>com.xu</groupId>
    <artifactId>lang-chain-pg</artifactId>
    <version>1.0.0</version>
    <name>lang-chain-pg</name>

    <description>Demo project for Spring Boot</description>

    <properties>
        <java.version>25</java.version>
    </properties>

    <dependencyManagement>
        <dependencies>
            <dependency>
                <groupId>dev.langchain4j</groupId>
                <artifactId>langchain4j-bom</artifactId>
                <version>1.6.0</version>
                <type>pom</type>
                <scope>import</scope>
            </dependency>
        </dependencies>
    </dependencyManagement>

    <dependencies>

        <!--Spring Boot 的 Web starter-->
        <dependency>
            <groupId>org.springframework.boot</groupId>
            <artifactId>spring-boot-starter-web</artifactId>
        </dependency>

        <!--LangChain4j 的 Spring Boot starter-->
        <dependency>
            <groupId>dev.langchain4j</groupId>
            <artifactId>langchain4j-spring-boot-starter</artifactId>
        </dependency>

        <!--LangChain4j 针对 OpenAI 的 Spring Boot starter-->
        <dependency>
            <groupId>dev.langchain4j</groupId>
            <artifactId>langchain4j-open-ai-spring-boot-starter</artifactId>
        </dependency>

        <!--LangChain4j 对 PostgreSQL 向量扩展（pgvector）的支持库-->
        <dependency>
            <groupId>dev.langchain4j</groupId>
            <artifactId>langchain4j-pgvector</artifactId>
        </dependency>

        <!-- 本地模型 -->
        <dependency>
            <groupId>dev.langchain4j</groupId>
            <artifactId>langchain4j-embeddings-bge-small-zh-v15-q</artifactId>
            <version>1.7.1-beta14</version>
        </dependency>

        <!-- 文档解析 -->
        <dependency>
            <groupId>dev.langchain4j</groupId>
            <artifactId>langchain4j-document-parser-apache-tika</artifactId>
        </dependency>

        <!--Spring Boot 的开发工具，提供热部署、自动重启等功能，加速开发过程-->
        <dependency>
            <groupId>org.springframework.boot</groupId>
            <artifactId>spring-boot-devtools</artifactId>
            <scope>runtime</scope>
            <optional>true</optional>
        </dependency>

        <!--简化 Java 代码的工具库-->
        <dependency>
            <groupId>org.projectlombok</groupId>
            <artifactId>lombok</artifactId>
            <optional>true</optional>
        </dependency>

        <!--Spring Boot 的测试 starter-->
        <dependency>
            <groupId>org.springframework.boot</groupId>
            <artifactId>spring-boot-starter-test</artifactId>
            <scope>test</scope>
        </dependency>

    </dependencies>

    <build>
        <plugins>
            <plugin>
                <groupId>org.apache.maven.plugins</groupId>
                <artifactId>maven-compiler-plugin</artifactId>
                <configuration>
                    <annotationProcessorPaths>
                        <path>
                            <groupId>org.projectlombok</groupId>
                            <artifactId>lombok</artifactId>
                        </path>
                    </annotationProcessorPaths>
                </configuration>
            </plugin>
            <plugin>
                <groupId>org.springframework.boot</groupId>
                <artifactId>spring-boot-maven-plugin</artifactId>
                <configuration>
                    <excludes>
                        <exclude>
                            <groupId>org.projectlombok</groupId>
                            <artifactId>lombok</artifactId>
                        </exclude>
                    </excludes>
                </configuration>
            </plugin>
        </plugins>
    </build>

</project>

2 配置

spring:
  application:
    name: lang-chain-rag

3 代码

1 RagConf

package com.xu.conf;

import dev.langchain4j.data.segment.TextSegment;
import dev.langchain4j.model.embedding.EmbeddingModel;
import dev.langchain4j.model.embedding.onnx.bgesmallzhv15q.BgeSmallZhV15QuantizedEmbeddingModelFactory;
import dev.langchain4j.store.embedding.EmbeddingStore;
import dev.langchain4j.store.embedding.pgvector.PgVectorEmbeddingStore;
import org.springframework.context.annotation.Bean;
import org.springframework.context.annotation.Configuration;

@Configuration
public class RagConf {

    @Bean
    public EmbeddingModel embeddingModel() {
        //return new AllMiniLmL6V2EmbeddingModelFactory().create();
        // 使用 BGE-small-zh-v1.5-quantized 模型
        return new BgeSmallZhV15QuantizedEmbeddingModelFactory().create();
    }

    @Bean
    public EmbeddingStore<TextSegment> embeddingStore() {
        return PgVectorEmbeddingStore.builder()
                .dropTableFirst(true)
                .createTable(true)
                .table("embedding")
                .user("postgres")
                .host("localhost")
                .password("123456")
                .database("postgres")
                .port(5432)
                .dimension(512)
                .build();
    }

}

2 RagController

package com.xu.controller;

import dev.langchain4j.data.document.DocumentSplitter;
import dev.langchain4j.data.document.loader.FileSystemDocumentLoader;
import dev.langchain4j.data.document.parser.apache.tika.ApacheTikaDocumentParser;
import dev.langchain4j.data.document.splitter.DocumentByWordSplitter;
import dev.langchain4j.data.segment.TextSegment;
import dev.langchain4j.model.embedding.EmbeddingModel;
import dev.langchain4j.store.embedding.EmbeddingSearchRequest;
import dev.langchain4j.store.embedding.EmbeddingStore;
import dev.langchain4j.store.embedding.EmbeddingStoreIngestor;
import lombok.AllArgsConstructor;
import org.springframework.http.ResponseEntity;
import org.springframework.util.CollectionUtils;
import org.springframework.web.bind.annotation.GetMapping;
import org.springframework.web.bind.annotation.RequestMapping;
import org.springframework.web.bind.annotation.RequestParam;
import org.springframework.web.bind.annotation.RestController;

import java.util.HashMap;
import java.util.Map;
import java.util.stream.Collectors;

@RestController
@AllArgsConstructor
@RequestMapping("/rag")
public class RagController {

    private final EmbeddingModel embeddingModel;

    private final EmbeddingStore<TextSegment> embeddingStore;

    /**
     * RAG数据入库
     *
     * @return 数量
     */
    @GetMapping("/save")
    public Object chat() {
        // 1️⃣ 使用 Tika 解析 docx/pdf 等文件
        var documents = FileSystemDocumentLoader.loadDocuments(
                "D:\\SourceCode\\简历",
                new ApacheTikaDocumentParser()
        );
        // 2️⃣ 定义自定义的文本拆分器 chunkSize=100 表示每段最大 500 tokens，overlap=30 表示重叠 30 tokens
        var splitter = new DocumentByWordSplitter(100, 30);
        // 3️⃣ 构建带自定义拆分器的 ingestor
        var ingestor = EmbeddingStoreIngestor.builder()
                .documentSplitter(splitter)
                .embeddingStore(embeddingStore)
                .build();
        // 4️⃣ 执行嵌入生成与存储
        ingestor.ingest(documents);
        return documents.size();
    }

    /**
     * RAG数据查询
     *
     * @param query 查询
     * @return 结果
     */
    @GetMapping("/search")
    public Object search(@RequestParam String query) {
        var searchRequest = EmbeddingSearchRequest.builder()
                .queryEmbedding(embeddingModel.embed(TextSegment.from(query)).content())
                .maxResults(5)
                .minScore(0.6)
                .build();

        var matches = embeddingStore.search(searchRequest).matches();
        var results = matches.stream()
                .map(match -> {
                    Map<String, Object> map = new HashMap<>();
                    map.put("embeddingScore", match.score());
                    map.put("embeddingId", match.embeddingId());
                    return map;
                }).collect(Collectors.toList());

        return ResponseEntity.ok(results);
    }

}

4 测试