NVIDIA HPC Developer

🎉 "This is the 20th anniversary of CUDA. We have been working on this architecture for 20 years ... to now have built up hundreds of millions of GPUs and computing systems around the world that run CUDA."

This is a fantastic thread. AIMIP is rallying the community around open evaluations of climate models, and we’re proud to contribute.

High-resolution weather modeling was a major theme at #EGU26. Learn how to train your own regional diffusion models for downscaling and forecasting with NVIDIA Earth-2 and PhysicsNeMo. Watch to start building AI-powered local weather models on YouTube 👇 nvda.ws/4diUjrA

NVIDIA’s Dion Harris joined leaders across government, national labs and industry to discuss how full-stack accelerated computing can power a new era of American science. The Genesis Mission is about more than faster progress. It’s about entirely new possibilities. #AIExpoDC

Finally able to talk about what I've been heads-down on for 6 months at @nvidia 🦀⚡ We just open-sourced cuda-oxide — an experimental rustc backend that lets you write CUDA kernels in pure Rust. No DSLs. No FFI. No source-to-source step. Single source. Short🧵👇

What happens when you let LLM agents write the code and NVIDIA GPUs do the heavy lifting? 🤖⚙️ In a Kaggle churn competition, it meant 150 stacked models, 850 runs, and a gold medal—powered by NVIDIA cuDF, NVIDIA cuML, XGBoost, and PyTorch. 🏅 Our latest blog breaks down the four-step playbook you can reuse for your own tabular ML. nvda.ws/41PDSxM

CUDA Live: Scaling HPC with Multi-GPU Communication Libraries x.com/i/broadcasts/1…

🌍💚 Happy Earth Day. We are spotlighting five projects where AI is helping solve some of the Earth’s greatest challenges. Check them out below . . . 🧵👇 ⛈️ Earth-2 climate simulation 🦧 Endangered species conservation ♻️ Robots sorting recyclables 🌊 Tsunami warnings 🌌 Earth observation insights

🌍 Real-time data is the ultimate tool for Earth Day. NVIDIA Earth-2 models like HealDA, built with collaboration from @MITREcorp and @NOAA, turn satellite observations into life-saving forecasts in just minutes. We’re moving past physics modeling alone to on-demand AI that helps society stay one step ahead of storms.👇 #earth-2" target="_blank" rel="nofollow noopener">blogs.nvidia.com/blog/earth-day…

⚛️ Explore how AI physics can accelerate clean, modular nuclear reactor design. By leveraging NVIDIA CUDA-X libraries, PhysicsNeMo, and Omniverse libraries, see how nuclear developers address these challenges with GPU-accelerated digital twins. nvda.ws/4tmWYax

Build custom, GPU-native atomistic simulation workflows for chemistry and materials science with the new NVIDIA ALCHEMI Toolkit. 🧪🧬 See how @OrbitalHardware, MatGL and @matlantis_en have integrated their platforms with ALCHEMI and try it yourself with pip install nvalchemi-toolkit. ⚙️ Read more: nvda.ws/3OHp1lY

You no longer have to choose between the power of Slurm and the flexibility of Kubernetes. NVIDIA Slinky lets you run Slurm natively on Kubernetes — it enables topology‑aware scheduling for architectures such as the NVIDIA GB200 NVL72, supports nondisruptive rolling updates, and has been proven at a scale of 8,000+ GPUs in production. Technical deep dive 👇 developer.nvidia.com/blog/running-l…

🌌 How to scale your HPC performance? In our upcoming developer livestream our CUDA team will share how NVSHMEM enables multiple GPUs to work in parallel on large-scale AI training, simulations, and rendering tasks. They'll walk through a live demo of NVSHMEM, dive deep on API calls to launch CUDA kernels across sets of GPUs, and end with a Q&A session. 📅 April 23, 1:00 PM - 1:30 PM CDT 📌 Register: nvda.ws/3QsameS Get your questions ready (and share in comments). We will see you there. 🙌

✨ Just published: technical guide for NVIDIA Ising 🥇 World’s first family of open AI models for building quantum processors AI-powered calibration + error correction workflows Delivers up to 2.5x faster decoding and 3x higher accuracy⚡ Built for fault-tolerant, hybrid quantum-classical systems Includes open models, weights, datasets, and workflow “recipes” for customization 📚 developer.nvidia.com/blog/nvidia-is…

Our new open-source AI model family will transform quantum computing by using agentic AI to solve it's most complex challenges, like real-time error correction. Nic Harrigan joins the NVIDIA AI Podcast to explain how NVIDIA Ising will help turn today's quantum hardware into tommorow's quantum-GPU supercomputers. 🌌 📺 Watch the full episode: youtube.com/watch?v=OFEY5-…

You won't want to miss any of the exciting Quantum Day announcements. Join us 📅 April 14, 8 AM - 1:30 PM Pacific. 🔗nvda.ws/3PTJVi6

👀First set of open-sourced AI models to accelerate the path to useful quantum computers. 👇

👀 Cut checkpoint costs with about 30 lines of Python and nvCOMP. At scale, storage, checkpointing, and data movement can become major sources of inefficiency. 💸 NVIDIA nvCOMP uses GPU-accelerated, lossless compression to reduce checkpoint size, lower storage overhead, minimize GPU idle time, and accelerate data movement across AI training workflows, all with minimal code changes. Technical blog 👇 developer.nvidia.com/blog/cut-check…