Spring AI is a project that brings AI engineering patterns and practices to Spring Boot developers, offering clean and idiomatic abstractions. The new release, m4, builds upon Spring Boot 3.4 and adds several features, including support for Amazon Bedrock Convertse, more expressive function calling in Java and Kotlin, and the first cut of support for advanced and modular RAG. RAG refers to using data from a system or service to inform the response generated by a chat model. The release also includes updates to vector store integrations and chat models, as well as a comprehensive chat model comparison page in the documentation. Spring AI aims to make connecting AI models with data and business logic easy, focusing on integration rather than building models from scratch. The release includes new support for describing and invoking functions, allowing models to ask questions about systems and services. A simple example demonstrates the definition of a FunctionCallback and the Spring AI ChatClient, which is used to interact with a Spring AI ChatModel. The example shows how to define a FunctionCallback that returns the weather for a given city, and how to use it in a test to verify the expected response. The test verifies that the model can ask questions about the system and services, and that the function callback provides the expected response. Overall, Spring AI makes it easy to integrate AI models with data and business logic, and the new release adds several features to support this goal.