On February 6, 2026, the TornadoVM team from UNIMAN gave a talk in Ticino, Switzerland!
The talk presented TornadoVM as a solution for accelerating Java applications on GPUs, CPUs, and FPGAs without writing any GPU code, addressing the challenges of GPU access from Java in the AI era. The speakers highlighted the key innovations of the TornadoVM runtime platform and demonstrated practical optimization techniques using its high-level API.
The session also showcased GPULlama3.java, a pure Java large language model (LLM) library accelerated with TornadoVM, followed by a live-coding demo in IntelliJ IDEA highlighting TornadoVM’s profiling and tooling ecosystem.
