Java News Roundup: OpenJDK JEPs, Jakarta EE, GraalVM, TornadoVM, Micronaut, Quarkus, JobRunr, Maven

This week’s Java news roundup for July 27th, 2026, from InfoQ covers OpenJDK JEPs targeted and proposed to target for JDK 28, the GA release of GPULlama3.java 1.0, point releases of Micronaut, Quarkus, and JobRunr, a maintenance release of JDKUpdater, the sixth release candidate of Maven 4.0, and the first milestone release of Jakarta Agentic AI 1.0.

On the platform side, OpenJDK activity around JDK 28 and the sixth Maven 4.0 release candidate sit next to runtime and framework updates: Micronaut, Quarkus, and JobRunr point releases, a JDKUpdater maintenance drop, and the GPULlama3.java 1.0 GA. Separately, Jakarta Agentic AI 1.0 reached its first milestone, extending the Jakarta EE line into agent-oriented APIs rather than only classic enterprise services.

For engineers and builders, the useful signal is concurrent movement across the stack: language and JDK targeting for 28, build tooling via Maven 4.0 RC6, application frameworks and background jobs via Micronaut, Quarkus, and JobRunr, plus an early Jakarta agentic AI surface. GPULlama3.java 1.0 GA adds a concrete GPU-oriented Llama path in Java that teams can evaluate against existing inference setups without leaving the JVM ecosystem.

Competitively, the roundup shows Java’s ecosystem still advancing on multiple fronts at once—OpenJDK, Maven, mainstream frameworks, and new AI-related Jakarta work—rather than a single product announcement. That pattern keeps Java in the same conversation as other stacks chasing agent tooling and local or GPU-backed model runtimes, while established frameworks keep shipping incremental releases.

Practical next steps: track which JEPs actually land for JDK 28, try Maven 4.0 RC6 in a non-production build, and review changelogs for the Micronaut, Quarkus, and JobRunr point releases before upgrading. Watch Jakarta Agentic AI 1.0 past this first milestone for API stability, and assess GPULlama3.java 1.0 against your hardware and model needs if you need Java-native GPU inference.