TornadoVM v4.0.0 is out!
✔ @Apple Metal backend → native Apple Silicon
✔ SIMD intrinsics for PTX → more performance
✔ @nvidia CUDA Graphs → reduced overhead
✅ SDKs available via SDKMAN! → https://t.co/hhGuKmr5tF
✅ Downloads at TornadoVM → https://t.co/8MZFy3HkAI
✅ Available on Maven Central → https://t.co/roD5gl7llB
✅ Docker images are available on DockerHub → https://t.co/lMUOKBMquK
Big thanks to our contributors! 👏
#Java #JDK25 #JDK21 #TornadoVM #opensource #AI #GPUs
Last week I was in Sofia for @jPrimeConf!
I presented our latest work on @TornadoVM and #GPULlama3 for seamless #GPU acceleration of local #AI inference with @QuarkusIO and @langchain4j.
Thanks to the organizers, speakers, attendees, and photographers for a great event.
Mary Xekalaki (@MXekalaki) from @csmcr will present this week our work at #ICPE 2026 in Florence.
“Pulse: A Profiling and Visualization Infrastructure for Heterogeneous Managed Systems”
📅 Date: Thursday, May 7
🕒 Time: 14:40–15:00
👉https://t.co/ihQOQtGGDt
@EU_HaDEA@_UKRO_
TornadoVM team at @DevoxxGR
Great discussions, insightful sessions, and a fantastic community pushing the boundaries of high-performance #Java and #GPUs.
📸 Photo credit: @dDoutsiopoulos
Last week we gave two presentations at @ocxconference and @DevoxxGR - and the recordings are now available:
🔹 TornadoVM: The Journey from Academia to Open Source Success
https://t.co/lafDcLlTxZ
🔹 Building and Running LLMs on GPUs Directly from Java with TornadoVM and https://t.co/5iLrNEuzxq
https://t.co/nivD16afVk
From open-source evolution to running LLMs on GPUs directly from Java - there’s a lot to explore.
Stay tuned for more!
#opensource #Java #TornadoVM #AI #LLM #GPUs
Excited to share that #TornadoVM is featured in the latest edition of @JAVAPROmagazin Magazine (April 2026), focused on "Java in the Age of AI."
👉 https://t.co/DgSFexvtgN
In this article, @mikepapadim shared our experiences on building #LLM inference libraries in #Java and accelerating them on #GPUs with #TornadoVM.
Great to see continued innovation in the Java ecosystem, especially around high-performance computing and AI acceleration.
#Java #community #AI #TornadoVM #GPUs
Airhacks TV episode 145 Q&A: BCE pattern from 1992 meets Java 25—Agent Smith (~130KB, zero dependencies) for multi-agent delegation, GPU Llama running local LLMs via TornadoVM, ZUnit test runner in ~300 lines. No Maven, no Gradle, just Java 25 👉 https://t.co/URRPZGqJT7 #airhacks #java #java25 #llm #airhacks
At Devoxx Greece today? Check out "Building and Running LLMs on GPUs Directly from Java with TornadoVM and https://t.co/7DhujOJK4Y" with Michalis Papadimitriou at 10:55-11:35 in MC 3.
https://t.co/QiQ3YlbMCl
#quarkusworldtour
The latest JAVAPRO special edition has landed and it’s all about Java in the Age of AI.
My latest article is included, where I dive into how we build GPU-accelerated libraries for LLM inference in pure modern Java using @tornadovm and https://t.co/pEHpA4oprX .
On my way to Brussels 🇧🇪 for #OCX2026
“@TornadoVM: From Academia to Open Source Success”
🕓 Tomorrow, 16:00–16:45 | 📍 Studio 4
👉 https://t.co/gEoPPBQyUw
Let’s connect if you’re there!
#OpenSource#Java#TornadoVM
Great to see external contributions driving real progress 🚀
Native #GPU support on Apple Silicon with Metal unlocked by #TornadoVM has now enabled running Devstral 2 models directly on-device — a big step forward for performance and accessibility.
This is exactly what makes #opensource so powerful 🔥
Great to see external contributions to https://t.co/pEHpA4oprX n @tornadovm@AdamBien added support for running Devstral 2 models on GPUs natively on Apple Silicon with Metal
https://t.co/YFQki4dSG9