In our previous blog post about Anthropic prompt caching, we explored how prompt caching dramatically reduces API costs and latency by reusing previously processed prompt content. We introduced Spring AI's five strategic caching patterns for Anthropic Claude models and showed how they automatically handle cache breakpoint placement while respecting the 4-breakpoint limit.

AWS Bedrock brings prompt caching to a broader ecosystem—supporting both Claude models (accessed via Bedrock) and Amazon's own Nova family. If you're considering Bedrock or already using it, you'll find the same Spring AI…

Reactive Support in Spring for Apache Pulsar Will Be Discontinued

TL;DR: We’ve decided to discontinue the Reactive support within the Spring for Apache Pulsar project.

Our team regularly evaluates our project portfolio with long-term sustainability in mind. When adoption declines, we thoughtfully retire projects / functionality to focus our efforts where the community needs them most. Based on an assessment of adoption metrics, download trends, and project activity, we’ve decided to remove the Reactive support within the project.

What does this mean for our users?

• The spring-pulsar-reactive module will be removed from all future releases, starting with Spring for Apache Pulsar 2.0.0.

• Spring Boot support of the Spring for Apache Pulsar Reactive components will be removed from all future releases of Spring Boot, starting with Spring Boot 4.0.0.

…

I'd like to introduce two new projects that are part of the Spring AI Community GitHub organization: Spring AI Agents, and Spring AI Bench. These two projects focus on using agentic coding tools—tools you likely already have in your enterprise.

In 2025 AI coding agents have matured to the point that they need to be seriously considered for enterprise Java development and general SDLC tasks. CLI Tools like Claude Code, Google’s Gemini CLI, Amazon Q Developer, and OpenAI’s assistants are examples from leading large AI labs, but there are also smaller startups and open-source options. These…

Continuing our Road to GA series, this week we're exploring the modularization effort happening with Spring Boot 4.

When Spring Boot 1.0 was released in 2014, it shipped with a single spring-boot-autoconfigure jar weighing in at 182 KiB. Of course, that initial version didn't support a great deal, but over the years, that has changed.

One of Spring's greatest strengths is the sheer number of technologies that it supports, but each new technology brings a cost. Each time we support something new, the autoconfigure jar grows. With Spring Boot 3.5, that single spring-boot-autoconfigure jar is now…

On behalf of the team and everyone who has contributed, I am pleased to announce the 4.32.1 release of the Spring Tools for Visual Studio Code, Eclipse and Theia.

This is a maintenance release that includes 4 bugfixes. At the moment, there are no further maintenance releases planned for the Spring Tools 4.x line.

Detailed changes can be found in the release notes: https://github.com/spring-projects/spring-tools/releases/tag/4.32.1.RELEASE

downloads

To download the distribution for Eclipse and find links to the marketplace entries for Visual Studio Code and Theia, please go visit:

• Spring Tools: https://spring.io/tools/
…

Hi, Spring fans! Welcome to another installment of This Week in Spring! It's a wonderful tuesday here in my home town of San Francisco as I write this from my condo's balcony, fresh off more than three weeks on the road.

By the time we'll speak again in a week, Halloween will have come and gone. Are you all ready for Halloween? What're you gonna dress as? I always go as PHP code. The horror!

The excitement is building for Spring Boot 4 and Spring Framework 7, both of which are scheduled to be released in November of 2025. So fun! There's a ton of good stuff to look at this week, so let's dive…

Large language model API costs can accumulate quickly when applications repeatedly send the same prompt content. A typical scenario: you're building a document analyzer that includes a 3,000-token document in every request. Five questions about that document means processing 15,000 tokens of identical content at full price.

Anthropic's prompt caching addresses this by allowing you to reuse previously processed prompt segments. Spring AI provides comprehensive support through strategic caching patterns that handle cache breakpoint placement and management automatically.

In this blog post, we…

I am pleased to announce the availability of Spring Modulith 2.0 RC1, 1.4.4, and 1.3.10. The latter two ship bug fixes and the general dependency upgrades. The release candidate primarily contains polish of the new features introduced in the milestone releases and a few new ones:

• Support for application-module-specific Flyway migrations #1067
• Upgrade to Spring Boot 4.0 RC1 #1415

Find more details about the releases in the full changelog.

I am pleased to announce that the first milestone of Spring Shell 4.0 is now available on Maven Central!

What's new?

This milestone release is the first step toward a modern version of Spring Shell, aligned with the rest of the Spring portfolio.

The main focus of this first milestone is alignment with Spring Framework 7 and Spring Boot 4.

Spring Shell 4.0.0-M1 is now based on Spring Framework 7.0.0-RC2 and Spring Boot 4.0.0-RC1.

For the complete list of changes, please check the release notes.

What's next?

We plan to release Spring Shell 4.0 GA in November, after the Spring Boot 4.0 GA release…