🚀 Gemma4 Java runner:
• Single file, no deps
• E2B→31B + MoE
• GGUF + quant (F16→Q8)
• Vector API ⚡
• CLI + thinking modes
• GraalVM native + instant TTFT
Pure Java 😏
by @mukel.bsky.social
github.com/mukel/gemma4...
Our @cern.voxxeddays.ch 2026 talk is now live!
We introduce TornadoVM, and show how it’s already used in the ESA #GAIA mission.
Highlight: #GPULlama3, a #GPU accelerated #LLM inference in #Java, integrated with @langchain4j.dev + #Quarkus
👉 www.tornadovm.org/gpullama3
📽️ youtu.be/oRVtTEbtt_Y?...
TornadoVM v4.0.0 is out!
✔ Apple Metal backend → native Apple Silicon
✔ SIMD intrinsics for PTX → more performance
✔ CUDA Graphs → reduced overhead
✅ Downloads at TornadoVM → www.tornadovm.org/downloads
Big thanks to our contributors! 👏
#Java #JDK25 #JDK21 #TornadoVM #opensource #AI #GPUs
ICYMI our @cern.voxxeddays.ch 2026 talk with @maryxek.bsky.social on @tornadovm.org #Java #GPUs #AI and all👇
youtu.be/oRVtTEbtt_Y?...
We’re excited to see TornadoVM featured at OCX 2026.
Thanos Stratikopoulos will share how TornadoVM evolved from academic research into an open source success story.
🕓 16:00-16:45
📍 Studio 4
www.ocxconf.org/event/2026/a...
#OCX2026 #TornadoVM #opensource #Java #HPC #research
New paper accepted at #ICPE2026!
We present TornadoVMPulse, a profiling & visualization tool for #TornadoVM to analyze and optimize #Java apps on #GPUs.
Results for #GPULlama3:
⚡ 18.4% faster execution
📉 38% lower data-transfer cost
🇪🇺 @aero-project.eu
📍 Florence, May26
zenodo.org/records/1895...
The latest version of #LangChain4J supports our latest #GPULlama3 version 0.4.0 for native GPU acceleration of #Java 25 AI Inference!
Join Mary and Michalis at #NVIDIA #GTC to discover how to achieve GPU accelerated Inference for #Java enterprise!
TornadoVM at NVIDIA GTC (Mar 16–19, San Jose)
Session: “Build AI Libraries for NVIDIA GPUs in Pure Modern Java With TornadoVM” [P81211]
👉 www.nvidia.com/gtc/session-...
See how Java runs AI inference on NVIDIA GPUs.
Come meet us and discover more!
#opensource #Java #GPU #AI #TornadoVM
Great to see the collaboration moving forward in the @aero-project.eu - and proud to represent the University of Manchester and the @tornadovm.org team. Looking forward to productive discussions over the coming days.
🚀 🚀 Another major release for TornadoVM now with both JDK21 and 25 support!
You can start running your GPU accelerated AI inference workloads in Java today on all major JDK distros!
TornadoVM v3.0.0 is out!
What a milestone! This release brings:
✅ JDK 25 (25.0.2) compatibility
✅ Continued JDK 21 (LTS) support
New SDKs are now available via SDKMAN! & the #TornadoVM downloads page.
Great teamwork from the community 👏
#Java #JDK25 #opensource #Java #AI
TornadoVM v3.0.0 is out 🎉
✅ JDK 25 (25.0.2) compatible
✅ JDK 21 (LTS) fully supported
• Available via SDKMAN! & TornadoVM downloads page
• Docker images coming soon
👉 github.com/beehive-lab/...
👉 www.tornadovm.org/downloads
Thanks to all contributors! 👏
#Java #JDK25 #JDK21 #TornadoVM
Our #TornadoVM talk from Thanos Stratikopoulos (@stratika.bsky.social) at #VoxxedDaysTicino #VDT26 is now live on YouTube!
👉 youtu.be/2peFTUktkuY?...
1. From #Java to #GPUs without writing GPU code.
2. TornadoVM SDK installation via SDKMAN!
3. Featuring #GPULlama3.java + live #IntelliJ IDE demo.
Last week at #VoxxedDaysTicino, I presented our work on @tornadovm.org.
Great conversations, inspiring people - and the chance to admire Lugano🇨🇭
Big thanks to the organizers, and a special shout-out for the tour of the #Alps Supercomputer @cscsch.bsky.social 🙌
#opensource #Java #GPU #AI
Very excited for our upcoming talk at @cern.voxxeddays.ch with @maryxek.bsky.social on @tornadovm.org!
Join us to discover how to program production ready #Java applications on GPUs for the #AI era!
vdc26.voxxeddays.ch/talk/tornado...
🚀 Java meets GPU acceleration - see you on the road!
Over the next two months, our team will be sharing how our #opensource #Java technologies unlock #GPU acceleration for Java #AI at conferences across Europe and beyond.
Come chat with us, attend our talks and dive into the future of Java + GPUs.
Check out this #TornadoVM webinar given by Thanos Stratikopoulos last week, explaining how TornadoVM is used in our project to accelerate Java algorithms for the analysis of space data from the #GAIAMission.
Well worth a watch if you’re interested in #Java, #GPU & scientific computing.
In this @adoptium.net webinar, we showed how #TornadoVM runs seamlessly with Eclipse Temurin JDK 21, enabling the offloading of #Java compute-intensive workloads (e.g. #LLM inference) to #GPUs - without writing low-level GPU code.
👉 www.youtube.com/watch?v=qSyh...
@aero-project.eu
#opensource #AI
New Blog Post
How to configure TornadoInsight v1.4.2 and use it with TornadoVM SDK v2.0+
✅ Setup steps
✅ Runtime + performance insights
✅ Tested on Linux/macOS/Windows
👉 www.tornadovm.org/post/tornado...
#opensource #TornadoVM #Java #GPU @intellijidea.com
@aero-project.eu
🎄 Season’s greetings from TornadoVM!
Thank you to our contributors and community for another year of collaboration - and for helping us reach 10k commits along the way.
Wishing you a Merry Christmas and a Happy New Year ✨
TornadoVM SDKs are now available via SDKMAN!
👉Easier installation & version management
📝 Blog: www.tornadovm.org/post/getting...
Install it now: sdkman.io/sdks/tornado...
#opensource #Java #TornadoVM #AI #GPU
New blog post: TornadoVM SDK Quick Start (2.0+)
Learn:
• How to install TornadoVM SDK on Linux, macOS & Windows
• How to run tornado --devices
• How to use TornadoVM with Maven
👉 www.tornadovm.org/post/getting...
#opensource #Java #TornadoVM #GPUs
My talk at #OCX2026 is accepted!
I’ll share the journey of @tornadovm.org - from PhD research to a production-ready #OpenSource framework enabling #GPU acceleration and #AI in #Java, with impact in #EUResearch and @aero-project.eu.
www.ocxconf.org/event/2026/
#horizoneurope
#UKRI
The new #GPULlama3.java release showcases results from our EU-funded project, integrating @quarkus.io + @tornadovm.org to enable #GPU-accelerated #LLM inference in #Java!
🌐 gpullama3.dev
🔗 github.com/beehive-lab/...
#horizoneurope #opensource #AI
New release of #GPULlama3.java v0.3.0 🚀
Faster NVIDIA inference & pure Java GPU demo (@quarkus.io + @langchain4j.dev, no native dependencies).
Learn more: gpullama3.dev
GitHub: github.com/beehive-lab/...
A great achievement from our team!
#opensource #Java #AI #LLM #GPU #TornadoVM
🚀 gpullama3.java v0.3.0 is out!
Faster FP16 & Q8 inference on NVIDIA GPUs, smoother setup with the new TornadoVM SDKs and seamless Quarkus integration.
Also, here is a demo of @quarkus.io + @langchain4j.dev running GPU inference on NVIDIA through @tornadovm.org fully in Java, no native deps!
Another major release for #TornadoVM focused on developer productivity for #AI execution in #Java
Congratulations to the whole team!
Docker images now available for TornadoVM v2.0.0!
See here how to run:
👉https://github.com/beehive-lab/docker-tornadovm
These images support #NVIDIA #GPUs or #Intel #GPUs, making it easier than ever to get started with heterogeneous computing using #TornadoVM.
#containers #Java #acceleration
TornadoVM 2.0.0 release is out!
Major upgrades for performance, portability & user experience! 🚀
🔗 github.com/beehive-lab/...
Thanks to all our contributors 👏
#opensource #AI #LLM #Java #TornadoVM
