Jlama

Jlama Project

Project setup

To install langchain4j to your project, add the following dependency:

For Maven project pom.xml

<dependency>
    <groupId>dev.langchain4j</groupId>
    <artifactId>langchain4j</artifactId>
    <version>0.34.0</version>
</dependency>

<dependency>
    <groupId>dev.langchain4j</groupId>
    <artifactId>langchain4j-jlama</artifactId>
    <version>0.34.0</version>
</dependency>

<dependency>
    <groupId>com.github.tjake</groupId>
    <artifactId>jlama-native</artifactId>
    <!-- for faster inference. supports linux-x86_64, macos-x86_64/aarch_64, windows-x86_64 
        Use https://github.com/trustin/os-maven-plugin to detect os and arch -->
    <classifier>${os.detected.name}-${os.detected.arch}</classifier>
    <version>${jlama.version}</version> <!-- Version from langchain4j-jlama pom -->
</dependency>

For Gradle project build.gradle

implementation 'dev.langchain4j:langchain4j:0.34.0'
implementation 'dev.langchain4j:langchain4j-jlama:0.34.0'

Embedding

The Jlama Embeddings model allows you to embed sentences, and using it in your application is simple. We provide a simple example to get you started with Jlama Embeddings model integration. You can use any bert based model from HuggingFace, and specify them using the owner/model-name format.

Create a class and add the following code.

import dev.langchain4j.data.embedding.Embedding;
import dev.langchain4j.data.segment.TextSegment;
import dev.langchain4j.model.jlama.JlamaEmbeddingModel;
import dev.langchain4j.model.embedding.EmbeddingModel;
import dev.langchain4j.store.embedding.EmbeddingMatch;
import dev.langchain4j.store.embedding.EmbeddingStore;
import dev.langchain4j.store.embedding.inmemory.InMemoryEmbeddingStore;

import java.util.List;

public class HelloWorld {
    public static void main(String[] args) {
        EmbeddingModel embeddingModel = JlamaEmbeddingModel
                                        .modelName("intfloat/e5-small-v2")
                                        .build();

        // For simplicity, this example uses an in-memory store, but you can choose any external compatible store for production environments.
        EmbeddingStore<TextSegment> embeddingStore = new InMemoryEmbeddingStore<>();

        TextSegment segment1 = TextSegment.from("I like football.");
        Embedding embedding1 = embeddingModel.embed(segment1).content();
        embeddingStore.add(embedding1, segment1);
        
        TextSegment segment2 = TextSegment.from("The weather is good today.");
        Embedding embedding2 = embeddingModel.embed(segment2).content();
        embeddingStore.add(embedding2, segment2);
        
        String userQuery = "What is your favourite sport?";
        Embedding queryEmbedding = embeddingModel.embed(userQuery).content();
        int maxResults = 1;
        List<EmbeddingMatch<TextSegment>> relevant = embeddingStore.findRelevant(queryEmbedding, maxResults);
        EmbeddingMatch<TextSegment> embeddingMatch = relevant.get(0);

        System.out.println("Question: " + userQuery); // What is your favourite sport?
        System.out.println("Response: " + embeddingMatch.embedded().text()); // I like football.
    }
}

For this example, we'll add 2 text segments, but LangChain4j offers built-in support for loading documents from various sources: File System, URL, Amazon S3, Azure Blob Storage, GitHub, Tencent COS. Additionally, LangChain4j supports parsing multiple document types: text, pdf, doc, xls, ppt.

The output will be similar to this:

Question: What is your favourite sport?
Response: I like football.

Of course, you can combine Jlama Embeddings with RAG (Retrieval-Augmented Generation) techniques.

In RAG you will learn how to use RAG techniques for ingestion, retrieval and Advanced Retrieval with LangChain4j.

A lot of parameters are set behind the scenes, such as timeout, model type and model parameters. In Set Model Parameters you will learn how to set these parameters explicitly.

More examples

If you want to check more examples, you can find them in the langchain4j-examples project.