langchain4j

Why we included this project

Java teams that want LLM features inside their existing codebase are the audience here, and they don't have to stand up a separate Python service to get them. LangChain4j wraps model providers and vector stores behind one API, so swapping OpenAI for Ollama or moving between embedding stores is a configuration change rather than a rewrite. It also bundles the everyday pieces: prompt templating, chat memory, tool calling with MCP support, and higher-level agent and RAG patterns, packaged to fit naturally into Quarkus and Spring Boot. The bundled examples take you from an empty project to a working prototype quickly, which makes it a reasonable foundation for enterprise Java shops adding conversational or retrieval-augmented features.