MONAILabel 0.8.5 on AMD Instinct™ GPUs: Interactive Medical-Image Annotation
rocm.blogs.amd.com — Learn how to run MONAILabel 0.8.5 for interactive medical-image annotation on AMD Instinct™ GPUs.
Infrastructure & Hardware
Infrastructure and hardware comprise the physical compute systems, specialized chips, and data center components that power artificial intelligence workloads. Truncated covers chip architectures, supply chains, energy requirements, and the hardware deployments that enable large scale computing.
Rafay’s playbook on helping neoclouds graduate beyond selling just infrastructure
www.sdxcentral.com — CPO explains how token routing and software nous keeps GPUs busy
AI’s next infrastructure constraint
www.datacenterdynamics.com — Data center expansions have long been represented in gigawatts, but measurements of terabits-second and microseconds are now entering the constraint conversation
Microsoft's Project Zenith Arrives On AMD Ryzen AI Halo, Bringing Ready-To-Code Windows 11 To 192GB AI Mini PCs
wccftech.com — Microsoft's Project Zenith arrives on AMD Ryzen AI Halo, bringing Windows 11 developer tools and local AI workflows to a powerful 192GB mini PC.
Expedera Demonstrates Edge AI Subsystem Pairing with Cadence
www.unite.ai — Expedera announced on November 9, 2026, that it had paired its Origin Evolution neural processing unit (NPU) IP with Cadence’s Tensilica NeuroEdge 130 AI Co-Processor in a pre-silicon simulation, demonstrating how the tw...
Riot Platforms signs 910MW agreement with unnamed frontier AI lab
www.datacenterdynamics.com — Capacity will be provided at Riot's Corsicana, Texas, data center
Digital Realty launches liquid cooling lab in São Paulo, Brazil
www.datacenterdynamics.com — Company's first DRIL lab in Latin America, follows launch of London lab earlier this year
Q.ANT partners with Ionos for photonic computing pilot program
www.datacenterdynamics.com — Successful applicants will be able to test and evaluate their AI models on Q.ANT’s photonic hardware
What hardware does your PC need to run a local AI model?
www.engadget.com — The kind of AI models you can run depends on your machine's power. Let's unpack what your everyday PC will need, and how much it will cost you.
Lawyer builds private AI server with eight used Nvidia V100 GPUs — up to 215 tokens per second to keep legal files offline, costing about $2,400 per four-GPU board
A little ❤️ in the cloud.
CoreWeave to Power Wayve's AI Driver Training in Multi-Year Agreement
investors.coreweave.com — Wayve will train its end-to-end foundation models on CoreWeave Cloud, expanding training and evaluation as it accelerates commercial deployment for autonomous vehicles CoreWeave , Inc. (Nasdaq: CRWV), The Essential Cloud for AI®, today announced a multi-year agreement with Wayve, the global leader in embodied intelligence building the AI Driver for autonomous vehicles. Wayve will train its foundation models on CoreWeave Cloud with NVIDIA Blackwell GPUs , expanding its AI infrastructure for training, evaluation, and simulation. Wayve began using Weights Biases and has now scaled into leveraging CoreWeave Cloud. Wayve’s AI Driver is becoming the driving intelligence layer the automotive industry is choosing to build on, from personally-owned cars to robotaxis. It learns to drive from real-world data collected in more than 70 countries, which makes it a frontier training workload with physics and safety attached. Models of that kind demand purpose-built cloud solutions, with
GKE dynamic slicing on TPUs
cloud.google.com — Compared to creating a static nodepool for each TPU slice, GKE dynamic slicing reduces AI workload startup and workload recovery times.
A used RTX 3090 is still the best GPU for local AI, and it beats most new cards on value
www.xda-developers.com — This old GPU is still worth considering for local AI.
Google Escalates Nvidia Rivalry With New Versions of Its TPU Systems
Microsoft winds down 2023 compute deal with Oracle
www.datacenterdynamics.com — Newer contracts remain ongoing
Distributed LLM Inference on Kubernetes with llm-d, Wasm, & TornadoVM
www.infoq.com — In this episode, Edoardo Vacchi discusses the evolving landscape of WebAssembly on the backend, running Java workloads on GPUs using TornadoVM, and optimizing distributed AI inference using llm-d and vLLM on Kubernetes. Edoardo explains how llm-d manages cache locality (KV and prefix caches), request routing, and batch inference across Kubernetes clusters to maximize expensive GPU utilization.
AWS Launches TorchNeuron to Run PyTorch Models on AWS Trainium Without Modifications
www.infoq.com — At the recent PyTorch Conference North America 2026, Amazon introduced TorchNeuron, a native PyTorch backend that enables developers to run existing PyTorch code on AWS Trainium AI chips with virtually no code modifications.
Israel plans to build its own AI infrastructure with 300,000 GPUs
www.calcalistech.com — National AI chief Erez Askal says the country must control the technology powering critical services as energy, water and healthcare become increasingly dependent on artificial intelligence.
Apple’s $2,899 MacBook PROLED hides a massive unified memory bottleneck — why I think power users should abandon macOS for Nvidia RTX Spark and AMD Strix Halo
Dismiss email signup
Subscribe to 
The top AI news in your email delivered each morning. Choose categories