Normal view

There are new articles available, click to refresh the page.
Before yesterdayMain stream

NVIDIA CUDA 13.4 Packs CUDA Support For Windows-on-Arm Ahead of RTX Spark Launch, While Giving Devs Early Access To Vera Rubin Too

9 September 2026 at 22:10

NVIDIA CUDA Can Now Directly Run On AMD's RDNA GPUs Using The "SCALE" Toolkit 1

NVIDIA has released its latest CUDA 13.4 Toolkit, which adds support for Windows-on-Arm for RTX Spark PCs & enables early dev access to Vera Rubin. NVIDIA CUDA Adds The Final Touches Ahead of RTX Spark Next Month As CUDA 13.4 Gets Windows-on-Arm Support Today, NVIDIA is releasing its latest CUDA platform, the CUDA 13.4 Toolkit, which brings support for newer platforms while broadening the platform with new developer tools. The following are some of the main highlights of today's release: First and most important, the CUDA 13.4 toolkit brings CUDA support to Windows-on-Arm, which helps developers prepare Windows Arm64 applications […]

Read full article at https://wccftech.com/nvidia-cuda-13-4-support-windows-on-arm-ahead-of-rtx-spark-launch/

Cerebras CS-4 Delivers A 30x Uplift In AI This Year, But Next-Gen Racks Already In Works With CS-5 Pumping Out 10K TPS Per User In 2027 & CS-6 Bringing 3D Wafer-Scale SRAM

25 August 2026 at 21:40

A Cerebras server tower is shown with the Cerebras logo next to text saying 'CS-4 CS-5 CS-6' and 'The fastest gets faster.'

Cerebras is rolling out its CS-4 AI rack-scale solution this year but is also working on its next-gen CS-5 and CS6 solutions. Cerebras Takes Wafer Scale Engine To The Next-Level With CS-4, CS-5, and CS-6 AI Racks Last week, Cerebras took the curtains off its CS-4 rack-scale solution, which is powered by the WSE-3T chip. The WSE-3T is a boosted version of the WSE-3 (Wafer Scale Engine), offering much higher capabilities. At Hot Chips 2026, Cerebras is providing a deeper dive into its rack-scale solutions while also giving us a look at its next-gen solutions. So starting off with the […]

Read full article at https://wccftech.com/cerebras-cs-4-30x-uplift-ai-2026-next-gen-rack-solutions-cs-5-10k-tps-2027-cs-6-3d-wafer-scale-sram/

NVIDIA Vera Rubin Records A Massive 30x Increase In Throughput Per Watt Than Blackwell While Offering a 35x Token Cost Reduction Across Agentic AI Workloads

24 August 2026 at 17:10

NVIDIA Confirms Vera Rubin Launch In Q3 With Volume Ramp by Q4, As Blackwell Continues To See Massive Demand

NVIDIA's Vera Rubin platform offers disruptive token throughput at a much lower cost than Blackwell, showcasing its Agentic AI prowess. NVIDIA Software Stack Optimizations Continue To Propel Blackwell, But Vera Rubin Sits In A Whole Different League For Agentic AI Use Cases The latest on-silicon agentic AI performance results were measured by NVIDIA using real-world agentic coding trajectories. The benchmark used was SemiAnalysis AgentX, which evaluates AI infrastructure such as Vera Rubin in agentic-coding inference workloads across various models such as Kimi K3, MiniMax M3, GLM5.3, Qwen3.5, and DeepSeek V4 Pro. The following table explains the key metrics of AgentX: […]

Read full article at https://wccftech.com/nvidia-vera-rubin-30x-increase-in-throughput-per-watt-vs-blackwell-35x-token-cost-reduction-agentic-ai/

SpaceXAI Will Use NVIDIA Vera CPUs To Power Its Next-Gen Agentic AI Workloads, Also Bringing Vera Rubin Acceleration To Grok & Starmind AI Satellite

24 August 2026 at 15:05

Two men smiling and exchanging a beige box in an indoor setting, with one man wearing a black leather jacket and the other in a black T-shirt with 'Starbase' printed on it.

NVIDIA's Vera CPUs & Vera Rubin servers will be powering multiple SpaceXAI platforms such as its next-gen Agentic AI workloads, Grok AI, & Starmind AI satellite. SpaceXAI Doubles Down on NVIDIA's Latest AI Platforms As It Leverages Vera CPUs & Vera Rubin Servers To Accelerate Agentic AI Workflows, & Take AI Beyond The Terrestrial Boundaries Today, NVIDIA and SpaceXAI are making a blockbuster announcement in which both companies will come together to accelerate next-gen Agentic AI workloads further by leveraging Vera Rubin and Vera CPU platforms. These announcements come a few weeks after Elon Musk committed exclusively to NVIDIA GPUs […]

Read full article at https://wccftech.com/spacexai-to-use-nvidia-vera-cpu-vera-rubin-servers-agentic-ai-grok-starmind-ai-sattelite/

Intel Crescent Island GPUs Pack Up To 32 Xe3P Cores, Optimized For Agentic AI With Low-Cost LPDDR5X That Reaches Up To 480 GB Capacity

24 August 2026 at 15:00

The image showcases an Intel GPU code-named Crescent Island with '350W Air cooled PCIe,' featuring 'LPDDR 5x Memory' up to 480 GB, '32 Xe Cores,' and an open software stack for agentic AI, highlighted at Hot Chips 2026.

Intel has given the rundown on its next-gen AI inference accelerator, Crescent Island, which packs 32 Xe3 cores and 480 GB of LPDDR5X memory. Intel To Enter Agentic AI Accelerator Space With Xe3P-Powered Crescent Island GPUs, Offering Low-Cost & Low Power Through LP5X Memory & 350W TDPs The Intel Crescent Island GPU is based on the brand-new Xe3P architecture, which is the same graphics architecture that was teased by the company during its Panther Lake and Xe3 deep dives. The new silicon features of Crescent Island include: Software innovations for Crescent Island include: The new GPU architecture will be a […]

Read full article at https://wccftech.com/intel-crescent-island-gpus-32-xe3p-cores-for-agentic-ai-low-cost-lpddr5x-up-to-480-gb/

NVIDIA Enters Full Production of Groq 3 LPX AI Inference Accelerator Chips, Supercharging Vera Rubin With The Fastest Token Generation Speeds Ever Recorded

24 August 2026 at 15:00

A close-up image of the AMD MI300X GPU, featuring its intricate circuit design against a black background.

NVIDIA's Groq 3 LPX is now in full production, offering big token generation speedups for Vera Rubin platforms in the Agentic AI space. NVIDIA Dials Up Vera Rubin NVL72 Token Generation Capabilities, Recording 3,400 TPS With Groq 3 LPX AI Inference Accelerators As part of its Hot Chips 2026 announcements, NVIDIA today announced that its Groq 3 LPX AI inference accelerator chip is in full production. This announcement follows the mass production announcements of Vera CPUs and Vera Rubin servers, marking the robust execution of NVIDIA's AI roadmap. The Groq 3 LPX racks serve as an extension to the NVIDIA […]

Read full article at https://wccftech.com/nvidia-groq-3-lpx-ai-inference-accelerator-full-production-supercharging-vera-rubin/

NVIDIA’s Vera Rubin Production Ramp-Up Is Now Squeezing TLC NAND Supply, Driving 512Gb Spot Prices To $21 After The June Slump

15 August 2026 at 16:42

A close-up of a Samsung semiconductor chip with visible circuitry patterning and the code 'KABBGGC'.

Such is the scale of NVIDIA's ongoing production ramp-up for its Vera Rubin platform that it's now manifesting itself in seemingly sparsely related nooks and cranies of the supply chain, such as the spot market for TLC NAND. In hindsight though, the rise in spot TLC prices is entirely reasonable, especially given the Context Memory eXtension (CMX) that NVIDIA is bringing onboard with the Vera Rubin platform. NVIDIA's production ramp-up of its Vera Rubin platform is squeezing TLC NAND supply by locking in a growing number of NAND cells within the CMX, and as enterprise SSD demand scales up in […]

Read full article at https://wccftech.com/nvidias-vera-rubin-production-ramp-up-is-now-squeezing-tlc-nand-supply-driving-512gb-spot-prices-to-21-after-the-june-slump/

NVIDIA’s Optics Partners Crush CPO Delay Rumors, With Coherent’s Scale-Out Revenue Starting Late 2026

14 August 2026 at 12:20

An NVIDIA chip labeled 'GH200 H100 NVL' is positioned centrally within a complex cooling system.

NVIDIA's partners on co-packaged optics, or CPO, have said that the deployment of the first server solutions remains on track. NVIDIA's CPO (Co-Packaged Optics) Partners Say All Is Well, As Delay Rumors Are Blown Away In The Dust Recently, there have been several rumors regarding the adoption of CPO in the data center ecosystem. As per the rumor mill, the adoption of CPO was said to be delayed for years, but during recent investor calls, both Lumentum and Coherent have stated that the production of CPOs remains on track. According to Lumentum, the demand for ultra-high power CPO lasers is […]

Read full article at https://wccftech.com/nvidia-optics-partners-crush-cpo-delay-rumors-with-coherents-scale-out-revenue-starting-late-2026/

NVIDIA Ditches AC Power For 800 VDC AI Factories To Scale Up Compute, Backed By Microsoft, Google and 80 Ecosystem Firms For 2H 2026

13 August 2026 at 06:25

A close-up view of a server rack with multiple Nvidia hardware components stacked vertically.

NVIDIA is spearheading the race to roll out its first 800 VDC platforms with the backing of Google & Microsoft in the second half of this year. NVIDIA's 800 VDC AI Factories Break Past The "Power Wall", Unleashing Scaled-Up Compute Performance By Effectively Cancelling AC Overheads As Datacenters continue to outstrip electricity generation, the need for more power-efficient & versatile solutions has become essential. This is why major companies like NVIDIA, Microsoft & Google have invested heavily in 800V DC or HVDC (High-Voltage Direct Current) infrastructure to power their upcoming AI platforms. The push towards DC over AC power is […]

Read full article at https://wccftech.com/nvidia-800-vdc-platforms-break-past-traditional-power-distros-to-scale-up-performance/

NVIDIA’s Kyber Racks To Sport 340.4TB Of DRAM, With HBM4E Priced At $19.76/GB And Each Rack Reportedly Priced At $41.6 Million

11 August 2026 at 18:44

An NVIDIA Kyber Side Car display unit is showcased at an exhibition with people gathered around it.

NVIDIA's upcoming Vera Rubin Ultra systems are in a state of flux at the moment, with persistent rumors pointing to a potential HBM de-spec, where the GPU giant might choose to lower the quantum of HBM4E within each GPU to preserve its margins. Well, the Bank of America (BofA) is out with an insightful report today, quantifying the dynamics between HBM4E and LPDDR5X-based SOCAMM2 modules within NVIDIA's upcoming Vera Rubin Ultra-based Kyber racks. NVIDIA will spend more on LPDDR5X than HBM4E when it comes to its upcoming Vera Rubin Ultra-based Kyber racks Currently, each base Rubin GPU is equipped with […]

Read full article at https://wccftech.com/nvidias-kyber-racks-to-sport-340-4tb-of-dram-with-hbm4e-priced-at-19-76-gb-and-each-rack-reportedly-priced-at-41-6-million/

NVIDIA Says It “Continually Optimizes Compute, Networking, Memory” To Deliver Best Performance & Efficiency To Its Customers Amidst Vera and Rubin Ultra “Downgrade” Rumors

7 August 2026 at 05:00

NVIDIA Quashes Rubin & Kyber Rack Delay Rumors, Says "Chip Roadmap Is Intact"

NVIDIA's Rubin Ultra & Vera chips are subject to various "downgrade" rumors, but the company says that it is part of its optimization strategy. NVIDIA Rubin Ultra AI Chips & Vera CPUs Feature Optimized Memory Configurations For Performance & Scale Recently, there has been lots of online chatter surrounding NVIDIA's Vera, Vera Rubin, and Rubin Ultra platforms. Most of the discussion is centered around rumors of a potential downgrade that will cut the memory subsystem on these chips significantly. Now, rumors surrounding downgrades, down-spec, and delays aren't new for NVIDIA chips. There have been similar stories in the past during […]

Read full article at https://wccftech.com/nvidia-optimizes-compute-memory-to-deliver-best-performance-efficiency-to-its-customers-amidst-vera-rubin-ultra-rumors/

AMD’s Helios Rack Reportedly Costs 40% More Than NVIDIA’s Vera Rubin, Yet Microsoft Hedges Its Bets By Committing To Deploy Both

29 July 2026 at 22:35

Satya Nadella in front of a large Microsoft logo on a blue background talking about AI.

Microsoft is moving aggressively to expand its compute footprint, and hedging its bets on the way by tapping rack-scale systems from both NVIDIA and AMD in what is an echo of its AI model-agnostic approach within Azure and Copilot. Microsoft's Satya Nadella has just declared that Azure will be among the first to deploy "rack scale AI infrastructure based on AMD Helios and NVIDIA Vera Rubin" Microsoft's CEO, Satya Nadella, has revealed a nugget during the just-concluded earnings call, declaring that Azure will be among the first cloud platforms to "deploy next generation rack scale AI infrastructure based on AMD […]

Read full article at https://wccftech.com/amds-helios-rack-reportedly-costs-40-more-than-nvidias-vera-rubin-yet-microsoft-hedges-its-bets-by-committing-to-deploy-both/

NVIDIA Trims Vera Rubin Memory as HBM4 Prices Threaten to Eat 29% of Every Rack’s Cost – Report

26 July 2026 at 04:19

A large, vertically standing, unbranded server rack filled with equipment is being installed by workers in a dimly lit data center.

NVIDIA Corporation is significantly reducing the memory used by its Vera Rubin NV72 rack-scale AI system in order to address the high prices and the ongoing memory shortage, according to an analysis from GF Securities. The courage followed after NVIDIA signed long term memory agreements ahead of the shortage in the memory industry which allowed it to weather the storm that most technology companies had to face due to the production constraints stemming from high demand. Reducing Memory Capacity Can Allow NVIDIA To Cut Rubin NVL72 Rack Costs By Nearly 50%, Says Report NVIDIA's Vera Rubin NVL72 racks are among […]

Read full article at https://wccftech.com/nvidia-trims-vera-rubin-memory-as-hbm4-prices-threaten-to-eat-29-of-every-racks-cost-report/

Intel Foundry Securing Packaging & Wafer Deal With NVIDIA To Make Next-Gen Feynman GPUs Could Be Its Biggest Customer Win Yet

24 July 2026 at 01:50

Two executives shake hands in front of a chip, with 'intel' and 'NVIDIA' logos prominently displayed in the background.

Intel Foundry might be heading into a big deal with NVIDIA, as the company is expected to offer packaging and wafer support for next-gen Feynman GPUs. NVIDIA Feynman GPUs Could Be Intel Foundry's Biggest Win To Date With Chipzilla Providing Both Packaging & Wafer Supply A deal between Intel and NVIDIA won't be a major surprise for those following our previous reports, but it is definitely a big win for Intel's Foundry business. Earlier this year, we reported that NVIDIA might be looking into Intel's upcoming process & advanced packaging technologies for its next-generation GPUs, codenamed Feynman. Well, according to […]

Read full article at https://wccftech.com/intel-foundry-nvidia-feynman-gpu-wafer-packaging-deal/

AMD Unleashes Instinct MI455X GPU, A 320 Billion Transistor Behemoth That Is Designed to Tackle NVIDIA’s Rubin With 50% More HBM4 Memory & Up to 40 PFLOPs of AI Compute

23 July 2026 at 18:25

AMD Instinct chip installed on a server motherboard.

AMD's Instinct MI455X GPUs extend the company's AI roadmap, bringing leading HBM4 capacities and over 40 PFLOPs of compute for Agentic AI, rivaling NVIDIA's Rubin chip. AMD Has An Answer To NVIDIA's Rubin, It's Called Instinct MI455X & It's An Engineering Marvel For Agentic AI With More HBM4 Than Any Other AI Chip On The Planet The AMD Instinct MI455X is the GPU that will power the Helios AI rack. This GPU is based on the latest CDNA 5 architecture and is made using TSMC's 2nm and 3nm process technologies. The chip packs 320 billion transistors, just 16 billion transistors […]

Read full article at https://wccftech.com/amd-instinct-mi455x-gpu-320b-behemoth-tackle-nvidia-rubin-with-432gb-hbm4-40-pflops-ai/

AMD EPYC Venice CPUs Stomp NVIDIA’s Vera With 20% Faster Single-Core & 2.2x Higher Throughput With Up to 256 “Zen 6” Cores, 203 Billion Transistors & Over 5 GHz+ Clocks

23 July 2026 at 17:45

A detailed close-up image of the Intel Core Ultra 9 185H processor die showcasing its intricate architecture and layout.

AMD has officially launched its EPYC Venice CPU family, spanning several chips, with performance leadership in Agentic AI with Zen 6 cores. The Industry's First 2nm HPC CPU Is Here, Meet EPYC Venice Family, Packing Up To 256 "Zen 6" Cores & 203 Billion Transistors One component that is really shaping up as the king of the Agentic AI era is the CPU. AMD is leveraging its brand new Zen 6 core architecture that will be used on its 6th Gen EPYC CPUs, codenamed Venice. AMD's EPYC Venice chips are the first HPC product to enter volume production on TSMC's […]

Read full article at https://wccftech.com/amd-epyc-venice-cpus-256-zen-6-cores-203b-transistors-over-5-ghz-stomp-nvidia-vera/

NVIDIA Aiming To Produce Up To 1000 Vera Racks Per Day After Shipping “Hundreds of Thousands” of Grace Standalone Servers As It Guns For Dominance In The AI CPU Market

22 July 2026 at 16:25

A series of NVIDIA data center GPUs is mounted on a server rack.

NVIDIA says that it will be able to make 1,000 Vera Rubin Racks per day following the success of its Grace CPUs in AI markets. Vera Will Be A Monumental Success For NVIDIA After Grace As The Firm Races To Build 1,000 Racks Per Day Since its inception, NVIDIA has been known as a GPU maker, but the company has slowly started to move away from that, and now recognizes itself as a full-stack system provider. That is why the company is making extra efforts to accelerate its CPU roadmap, and while Grace was its first full-on take on a […]

Read full article at https://wccftech.com/nvidia-to-produce-up-to-1000-racks-with-vera-cpus-per-day/

AMD Is Reportedly Pricing Its Helios Rack 40% Above NVIDIA’s Second-Gen Rubin, Confident That Customers Will Pay The Premium

21 July 2026 at 15:17

A large server rack filled with multiple network cables and computing hardware, viewed from a low angle in a dimly lit data center.

AMD is apparently no longer satisfied with being perceived as a perennial underdog, and is reportedly choosing to make a bold statement by pricing its Helios rackscale solution at a hefty premium to NVIDIA's Rubin. What's more, AMD has reportedly managed to rope in Microsoft as the product's first confirmed client, demonstrating the utility that Helios offers even at its supposedly elevated price point. Futurum believes AMD will price Helios at $5 million to $5.5 million per rack, versus an expected retail price of $3.5 million to $4 million for the second-gen NVIDIA Rubin rack As we explained in a […]

Read full article at https://wccftech.com/amd-is-reportedly-pricing-its-helios-rack-40-above-nvidias-second-gen-rubin-confident-that-customers-will-pay-the-premium/

❌
❌