TornadoVM 7.3 Released with Major Native GPU Acceleration Updates for Java

1 reports archived1 independent sourcesupdated 11/2/2026, 11:39:13

Synthesis & Latest Updates

1 Sources Cross-Validated

Open-source heterogeneous computing framework TornadoVM released version 7.3, refactoring its backend by removing JVMCI dependencies and adopting the FFM API to deepen integration with the Nvidia CUDA ecosystem. The update supports direct calls to cuBLAS/CUTLASS, multi-GPU execution, and hybrid APIs for Apple Metal, offering Java developers a more efficient acceleration solution for machine learning workloads.

LATEST/Open-source heterogeneous computing framework TornadoVM released version 7.3, refactoring its backend by removing JVMCI dependencies and adopting the FFM API to deepen integration with the Nvidia CUDA ecosystem. The update supports direct calls to cuBLAS/CUTLASS, multi-GPU execution, and hybrid APIs for Apple Metal, offering Java developers a more efficient acceleration solution for machine learning workloads.

Hourly heat curve

3 fully observed hours

Now

11.2

Peak

11.912:00

24h change

–

The line compares only sources observed throughout; gap hours are interpolated to keep the trend continuous.

Coverage timeline

Total 1 reports · Latest first

InfoQ 中文T2Nov 2, 11:39·68 pts
TornadoVM 7.3 Released with Major Native GPU Acceleration Updates for Java
- Architectural Refactor: Backend migrated from JNI to Foreign Function & Memory (FFM) API, removing experimental JVMCI dependencies and unifying processing paths across JDK 21-27.
- Ecosystem Integration: Deepened Nvidia CUDA support, allowing direct calls to cuBLAS and CUTLASS from TaskGraphs, plus added multi-GPU execution support.
- Cross-Platform Capability: Optimized compilation for OpenCL and Apple Metal backends, enabling seamless performance scaling from laptop development to server deployment.
View Related Reports & Analysis Original Source