🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

A great milestone from our team. First release of #GPULlama3.java. Going from Java to GPU acceleration via #TornadoVM! #opensource #Java #AI #LLM

We are working on a new big project for @GraalVM — Project Crema: github.com/oracle/graal/i…. It will lift Native Image's default closed-world assumption by allowing dynamic loading and execution of classes at run time. Stay tuned for updates!

A free guide to data structures & algorithms in Python: bit.ly/43kQr5S

Happy birthday to Web inventor & MIT Professor Emeritus Tim Berners-Lee! Things besides the WWW that didn’t exist before TBL: - Apple - Google - Facebook - YouTube - Instagram - Netflix - Uber - Spotify - & most technologies we use today Image v/Sotheby's

AI ❤️ TPUs

This is pretty cool: switching between NVIDIA Driver Module Flavors. docs.nvidia.com/datacenter/tes…

One of the best ways to reduce LLM latency is by fusing all computation and communication into a single GPU megakernel. But writing megakernels by hand is extremely hard. 🚀Introducing Mirage Persistent Kernel (MPK), a compiler that automatically transforms LLMs into optimized

Happy birthday to Alan Turing, the creator of modern computing. Image v/Fine Art Images & Heritage Images/age fotostock

Setting up a new Fedora 41 workstation in less than 2h with NVIDIA & Intel GPU drivers, libraries, #Java & TornadoVM! Easy-peasy. BTW, Fedora 42 does not currently run OpenCL (intel-compute-runtime seems to be broken), so, going back to Fedora 41.

The University of Manchester team extended Alfonso² Peterssen☕'s Llama3.java (github.com/mukel/llama3.j…) with GPU support via the GraalVM JIT and TornadoVM: infoq.com/news/2025/06/g… Nicely done!👏

Thanks to Alfonso² Peterssen☕ for the llama2.java port and Roger Allen (github.com/rogerallen/lla…) for the CUDA port, we were able to understand and shape TornadoVM for running LLMs and transformers architectures.

What's this? An abstract syntax tree? A bytecode? What does that even mean anymore? Next Wednesday, Octave Larose is going to explore those questions in his ICOOOLPS presentation. Our draft paper is online already: 2025.ecoop.org/details/ICOOOL… #interpreter #design #tradeoffs

New blogpost: How to enable NVIDIA Nsight Compute CLI in Fedora/Linux. 🔗jjfumero.github.io/posts/2025/07/… #profiling #cuda #configuration

That drum kit in the center 😍#tool

New TornadoVM release! v1.1.1 Check it out!

Learn how to use Numba to write CUDA kernels in pure python at our tutorial at WeAreDevelopers on July 9th in Berlin! wearedevelopers.com/world-congress…

Updating open WebUI to run local models with CUDA. 🔗gist.github.com/jjfumero/a3039…

Today's my last day at The University of Manchester! After almost 8 incredible years, it feels like a second home. Huge thanks to everyone for the amazing memories, especially developing #TornadoVM into a world-class open-source platform. So proud of what we achieved! Excited for what's next!

I am happy to share that I am starting a new position as a Senior Principal Software Developer Oracle , joining the Java Platform Group!

Starlink is currently in a network outage and we are actively implementing a solution. We appreciate your patience, we'll share an update once this issue is resolved.