Bootiful Spring Boot 3.4: Spring AI
I love Spring AI. It’s an amazing project designed to bring the patterns and practices of AI engineering to the Spring Boot developer. It’s got clean idiomatic abstractions that’ll make any Sring developer feel right at home, and it has a ton of integrations with all manner of different vector stores, embedding models, transcription models, image modes, and chat models.
The new release, m4, builds upon Spring Boot 3.4 and adds a ton of new features. As usual, I can’t hope to look at all the new features, but the release notes do an exquisite job.
- there’s new Amazon Bedrock Convertse support
- there’s been a ton of work done to support more expressive function calling, both in Java and Kotlin
- there’s the first cut of support for the ideas taking shape in the AI community around "advanced and modular RAG". RAG, of course, is retrieval augmented generation and refers to using data from a system or service to inform the response generated by a chat model. These ideas are articulated in this paepr and this one, and the building blocks are taking root in this new release. Amazing!
- …