📣 Dive into multi-node, multi-GPU computing with cuPyNumeric. Join our NVIDIA Connect With Experts Webinar and discover how to scale your #Python code effortlessly. Engage with top #developers and unlock the power of #CUDA-accelerated libraries. Register today. ➡️

🎉 NVIDIA cuDSS 0.6.0 is now available. This latest release introduces many usability, scalability, and performance features that deliver powerful new capabilities. Learn more about the new features and download the latest update. 👇 nvda.ws/3G1x88C

Want to share examples effortlessly or run #CUDA code directly in your browser? 🔎 Discover the power of Compiler Explorer for #CUDA development. Dive into our latest technical blog to learn how this incredible tool can streamline your CUDA workflow. 👇 nvda.ws/3HNE61F

📣 Now available, NVIDIA DOCA 3.0 including new features: ✅ InfiniBand Quantum-X800 + ConnectX-8 SuperNIC (GA) ✅ DOCA Argus Service: NIM container threat detection ✅ DPF: Trusted Host use-case (GA) ✅ SNAP Virtio-fs (Beta) ✅ Perftest RDMA benchmark tool (GA) Tech blog ➡️

📣 New release: NVIDIA Collective Communications Library 2.26. Updates include: ✅ PAT optimizations: CUDA warp-level optimizations improving PAT algorithm parallelism. ✅ Implicit launch order: Prevent deadlocks when a device is used simultaneously by multiple communicators and

Join us for #opensource & accelerated #Python tutorials, talks, sprints, and more. We have 14 talks, 2 poster sessions, and a special social at the Museum of Glass. 📅 July 7-13 📍 #SciPy2025 Register now ➡️ti.to/scipy/scipy2025

Prototype, fine-tune, and deploy with ease. 🙌 NVIDIA DGX Spark, your own personal AI supercomputer. ✅ 1 petaFLOP ✅ 128GB of coherent, unified system memory ✅ ConnectX-7 Smart NIC ✅ Up to 4TB Storage ✅ 150mm L x 150mm W x 50.5mm H Reserve yours now ➡️

GTC Highlights: Lighting Up Europe GTC Paris at VivaTech 2025 spotlighted how NVIDIA and partners are building the future of AI—from sovereign infrastructure to agentic systems. Jensen Huang’s keynote introduced major advances like Grace Blackwell systems in full production and

Add this SciPy tutorial to your calendar now ➡️ cfp.scipy.org/scipy2025/talk…

Discover the newly released #cuQuantum 25.06 with DMRG primitives for circuit simulations and 26x speedup in AI-driven QPU design with dynamic gradients. #QuantumComputing Learn more 👉 nvda.ws/4kvKsQW

⚡ Take GPU optimization to the next level. Our upcoming webinar will demystify CUTLASS and walk you through optimized implementation of GEMM kernel design on #NVIDIAHopper, with experts from NVIDIA and Colfax International. ➡️ Register now for the session that is best for your time

With native Python support now available, #CUDA brings #acceleratedcomputing to scientific #Python users. Join Christopher Lamb, NVIDIA’s VP of Software for Compute Platforms at #SciPy2025 to learn more. 🐍 nvda.ws/408pFvj

Millions of parts, one AI superchip. Go behind the scenes and watch how NVIDIA GB200 NVL72 is created, connecting 72 #NVIDIABlackwell GPUs into a single massive, liquid-cooled system that pushes the limits of innovation and collaboration. nvda.ws/4nDe69u

Curious about advanced CUDA kernel optimization techniques? 🤔 Our latest blog post delves into handwritten PTX code, another one of the many tools available to the expert CUDA programmer. 🔗 Read: nvda.ws/4ksIErN

🎉 cuda.cccl: GPU building blocks for #Python. Leverage cuda.cccl for efficient GPU computing with Pythonic interfaces to CUB and Thrust. Write custom algorithms without dropping into C++. Learn how to boost your Python GPU performance ➡️ nvda.ws/4lOMlJE

#NVIDIABlackwell is now generally available on Amazon Web Services. The NVIDIA GB200 NVL72 platform with NVIDIA NVLink, now available as Amazon EC2 P6e-GB200 instances, enables training and inference at scale to power AI breakthroughs. 🎥 See now to learn more ⬇️ nvda.ws/4eDefWq

What’s new in cuQuantum 25.06? Dynamic gradients, DMRG, and simulation speedup, and more. 25.06 updates all cuQuantum libraries: cuDensityMat, cuStateVec, and cuTensorNet. New features include gradients for quantum dynamics workflows, further optimizations for NVIDIA Grace

🌟 New webinar: learn how to accelerate real-world, HPC and AI enabled applications with #CUDA by efficiently harnessing the GPU performance. Boost your CUDA development skills and gain exclusive access to our developer resources, including tutorials, best practices, and starter

🌍 Our NVIDIA Earth-2 now offers subseasonal-to-seasonal weather forecasting with #AI, enabling large ensemble predictions at lower costs. Learn how this can benefit farmers, energy firms, and governments in our technical blog ➡️ nvda.ws/4lmewPY