quarkiverse/quarkus-langchain4j

Quarkus LangChain4j – Quick Overview

What it is: A set of Quarkus extensions that let you plug large‑language‑model (LLM) capabilities—via the LangChain4j library—directly into a Quarkus application. In practice, you add a Maven artifact, configure a provider (OpenAI, Ollama, Hugging Face, etc.), and you can call LLMs, work with embeddings, retrieve documents, or expose declarative AI services just like any other Quarkus bean.

Key capabilities

• Declarative AI services – define an interface and let the extension generate the LLM‑backed implementation.
• Multi‑LLM support – OpenAI GPT, Hugging Face models, Ollama, and others.
• Tool & embedding integration – use LLM‑driven tools, create/query vector embeddings.
• Document stores – connect to Redis, Chroma, Infinispan, etc., for RAG‑style retrieval.
• Native compilation – works with Quarkus native images for low‑footprint deployments.
• Observability – metrics and tracing via Quarkus’ built‑in stack.
• Pluggable auth – flexible authentication for the underlying model APIs.

Getting started Add the appropriate Maven dependency, for example:

<dependency>
  <groupId>io.quarkiverse.langchain4j</groupId>
  <artifactId>quarkus-langchain4j-openai</artifactId>
  <version>1.15.0.CR1</version>
</dependency>

Then configure your API key and use the generated AI service beans.

Where to learn more

• Full docs: https://docs.quarkiverse.io/quarkus-langchain4j/dev/index.html
• Sample projects and integration tests are in the repo under samples/ and integration-tests/.

Who it’s for Java developers building micro‑services with Quarkus who want to add LLM‑driven features (chat, summarisation, RAG, tool use) without wiring the LangChain4j library manually.

License: Apache 2.0.