HPE and NVIDIA are pushing liquid cooling as a strategic imperative for AI data centers, citing rack densities of 70 kW to 100 kW+ required for next-generation NVIDIA Blackwell GPUs that air cooling c…
HPE and NVIDIA are promoting their co-engineered HPE AI Factory with NVIDIA as a way for CIOs to build and deploy production-ready AI agents in clicks rather than months, using NVIDIA NIM Agent Bluepr…
Nearly 90% of IT leaders have already experienced security incidents tied to AI pilot programs, according to industry data cited in a CIO guidance piece from HPE and NVIDIA. HPE and NVIDIA outlined th…
HPE and NVIDIA are positioning Cray supercomputing architecture as the foundation of their joint AI factory portfolio, with the Leibniz Supercomputing Centre's upcoming Blue Lion system built on the H…
NVIDIA released DLSS 5, which introduces 3D-Guided Neural Rendering and granular developer controls for lifelike lighting and material detail, and is available now in NBA 2K27 from Visual Concepts and…
Andreessen Horowitz co-founder Ben Horowitz and Marc Andreessen are launching The Horowitz Andreessen Academy, a highly selective San Francisco school for young builders after high school or their fir…
NVIDIA released Isaac ROS 5.0 at the ROSCon conference in Toronto, Canada, adding agentic workflows, support for ROS Lyrical and Ubuntu 24.04, and new Isaac skills for setup and manipulation to its GP…
NVIDIA contributed a CUDA buffer backend to ROS 2 Lyrical that lets ROS 2 nodes exchange GPU-resident payloads through zero-copy transport when runtime conditions allow, and all nodes in NVIDIA Isaac …
Durabook, the rugged mobile solutions brand owned by Twinhead International Corporation, announced the Z14I-DX3 Rugged Mobile Laptop, a workstation-class edge AI platform with three integrated 14-inch…
Spectro Cloud added support for external inference endpoints to its PaletteAI Inference Launchpad, letting customers connect Amazon Bedrock alongside locally run open-weight models. The company demons…
NVIDIA published LatentMoE in January 2026, a mixture-of-experts architecture that performs expert arithmetic in a space four times narrower than the model's width and spends the saving on a larger ex…
A custom inference stack running Qwen 3.8 27B on NVIDIA B300 SXM6 GPUs reached 9,825 tokens per second on 8× B300 with DP4×TP2 plus Suffix4, and a maximum cluster ceiling of 16,931 tokens per second o…
IPronics announced a two-year strategic technology collaboration with the Barcelona Supercomputing Center – Centro Nacional de Supercomputación (BSC-CNS) to integrate the iPronics ONE programmable opt…
NVIDIA Dynamo-Triton release 26.07 now enables TensorRT multi-device inference, allowing a single TensorRT network to execute across multiple GPUs using NCCL-backed distributed collectives, with full …
Agent evaluation has shifted from scoring single function calls to scoring full tasks in a live execution environment, according to an analysis citing the Berkeley Function-Calling Leaderboard (BFCL),…
VLLM's PagedAttention and continuous iteration-level batching address the KV cache memory bottleneck that limits LLM inference throughput, according to a technical deep dive on the inference engine's …
Shrijith Venkatramana, an engineer building the LiveReview AI code review tool, published a technical explainer on LLM quantization showing that activation outliers concentrated in a handful of featur…
An engineer at g factor detailed how vLLM's PagedAttention and continuous iteration-level batching solve the memory-bandwidth bottleneck in production LLM inference, drawing on benchmarks run on dedic…
Mixture of Experts has changed the structure of AI inference serving by altering which tensors are active per token, what must remain close together, which transfers need strong local bandwidth, and h…
Unimore Racing, from the University of Modena and Reggio Emilia, won the Indy Autonomous Challenge at Laguna Seca on Sept. 3, the first IAC event in which autonomous racecars overtook one another on a…