Red Hat published a post arguing that 'the CPU is back' for LLM inference, citing an Intel and Georgia Tech paper that found CPU-side tool processing accounts for 50–90% of total latency in agentic wo…
Delta Electronics unveiled its GoCool-150 liquid-to-air coolant distribution unit at ASRock Rack's Computex 2026 booth, designed to discharge up to 150kW of heat from a single liquid-cooled server rac…
GMKtec's EVO-X3, a tower PC with AMD's Ryzen AI Max+ 395, 128GB unified memory, and an OcuLink port, was reviewed by Wendell of Level1Techs, who called it 'one of the fastest Strix Halo machines' he h…
Firebird, an emerging AI cloud, launched the CIS region's largest AI factory in Armenia, powered by NVIDIA accelerated computing and Dell Technologies infrastructure, with plans to deploy more than 70…
Prime Intellect's Prime Agent, a coding harness that replaces traditional tool menus with a live IPython kernel, can be run locally with DeepSeek V4 on NVIDIA's DGX Spark, and reportedly pushed Claude…
NVIDIA's NeMo Retriever enables a multimodal retrieval-augmented generation pipeline that extracts text offline with PDFium, detects page elements, tables, charts, and infographics via hosted NVIDIA N…
Jonathan Lorraine, a research scientist at NVIDIA's Learning and Perception Research group, focuses on AI for science, developing agents that propose, run, and verify their own experiments, and on mea…
China's enterprise AI market is moving beyond the model race, with 2025 data from the China Large Model Procurement Monitoring Report recording more than 7,500 large-model-related procurement contract…
Indosat Ooredoo Hutchison launched Zankore by Indosat on August 6, 2026, with Ooredoo Group, Nokia, and NVIDIA, targeting 1 GW of NVIDIA DSX AI Factory capacity in Indonesia, with construction in Bata…
At FMS 2026, SK hynix presented its next-generation memory architecture for the AI era, unveiling the first HBF standard specifications with Sandisk and showcasing AI memory solutions including CXL, D…
Red Hat launched asago, an open source project under the Apache 2.0 license, to automate translating AI governance policies into operational controls, generating safety tests and recommending guardrai…
Google Cloud announced that frontier AI lab Mirendil will use its AI Hypercomputer, including Google TPU v5P chips and NVIDIA AI infrastructure, for model pre-training and post-training. Mirendil is a…
NVIDIA has released Cosmos 3, an open physical AI foundation model family, under the Linux Foundation's OpenMDW 1.1 license, enabling teams to post-train models on their own data and hardware. The mod…
It's FOSS reports that Greg Kroah-Hartman has banned AI-generated patches from the drivers/staging area of the Linux kernel, while allowing them elsewhere, and GNOME has published a guide for bots to …
NVIDIA has brought its RTX Neural Texture Compression (NTC) technology to Windows-on-Arm, ahead of the general release of its RTX Spark PC platform. NTC, first demonstrated at GTC 2026 in May, can red…
SpaceX and NVIDIA announced a joint project to develop the Starmind AI1 satellite, powered by NVIDIA's Vera CPUs and Rubin GPUs, to process AI workloads in low-Earth orbit. SpaceX has petitioned the F…
NVIDIA's 45-page whitepaper on its Vera server CPU, built around the 88-core Olympus Arm v9.2 core, contains misleading marketing claims that undermine its strong hardware, according to an analysis. T…
Muse released Muse Code (beta), a terminal coding agent powered by its new Muse Spark 1.2 model, which features async background agents, a replay-exact runtime, and bundled skills for planning and exe…
The global computer microchip market is projected to nearly triple to $74.70 billion by 2035 from $30.03 billion in 2025, growing at 9.54% annually, according to SNS Insider data released via GlobeNew…
Microsoft Corp. disclosed that OpenAI accounts for approximately 70% of its AI sales, with OpenAI's revenue reaching about $24.1 billion on an annualized basis in fiscal 2026. The partnership, restruc…