llm-apps-java-spring-ai

Why we included this project

Java developers already working in Spring Boot can treat this as a working reference for adding LLM features without leaving the stack they know. It is a collection of runnable samples built on Spring AI, each covering a pattern that turns up in real applications: chat completions, retrieval-augmented question answering with PGVector, semantic search, structured output, tool calling, chat memory, and input/output guardrails. The same pattern is demonstrated against OpenAI, Ollama, and Mistral AI, which makes it easy to see how providers differ and to swap backends without touching your application code. There is also guidance on ingesting data from PDF, JSON, Markdown, and Tika, handy when you are assembling a knowledge base. Because the samples are grouped by use case, model, and pattern, you can jump to the one you need and adapt it to your own service.