AI Hardware News
Product launches, GPU shortages, and market shifts tracked since October 2025.
September 2026
OpenAI deepens chip cooperation with Samsung, possibly prepares to Double source AI ASICs from two foundries in a bid to get more in-house silicon to its data centers.
TSMC and Samsung will adopt ASML’s High NA EUV tools for advanced chipmaking as AI drives demand for smaller, more complex semiconductors.
Graphics card scams without the core and memory chips are spreading worldwide, with the latest report coming from Germany.
Nvidia’s $12.9 billion Hugging Face deal could protect its open AI ecosystem, deepen ties with developers and keep the platform from rivals.
Minisforum unveiled the NAS N5 Max-P495 and MS-S1 Max-P945 at IFA 2026. The NAS and mini-PC are powered by the AMD Ryzen AI Max+ Pro 495, which can be configured with up to 192GB of unified memory and packs a Radeon 8065S integrated GPU.
Most board partners expect the next-generation GPUs to launch toward the back half of 2027.
Nvidia expands beyond AI hardware with its $12.93 billion acquisition of Hugging Face, gains control of a major open AI model platform, vows to preserve its support for competing models, clouds, and hardware platforms.
Nvidia's Personal AI Router (PAIR) clustering utility lets agentic AI workloads take advantage of every spare GPU cycle on a home network, potentially making for faster execution and more private inference.
AMD's Gorgon Halo chips are breaking cover. Ahead of IFA, Lenovo showed off the THinkCentre X Ultra with an AMD Ryzen AI Max+ Pro 495.
DLSS 5 officially releases later this week, but we tested it on the RTX 5090 using community mods to see what kind of performance drop and extra power draw you can expect.
Nvidia invests $3.5 billion in MediaTek as the companies expand their partnership into custom AI infrastructure with NVLink Fusion, local AI computing, and automotive platforms.
August 2026
Nvidia's Q2 FY2027 revenue tops $96 billion as the company commits to buy up to $160 billion worth of memory, increasing its total commitments to $279 billion ahead of a major demand bump.
Nvidia reportedly agreed to buy open source AI platform Hugging Face for $12.9 billion as the chipmaker expands deeper into the AI ecosystem.
"Our demand is much greater than 70%," CEO Jensen Huang said on the earnings call.
SpaceX and xAI say Grok's agentic workloads will run on standalone Nvidia Vera CPUs, with a space-optimized Vera Rubin NVL72 rack carried by the Starmind satellite design from Q4 2027.
At Hot Chips 2026 Intel detailed Xeon 7 'Diamond Rapids', its 18A-P P-core server CPU: up to 256 cores per socket, 1.28 GB of last-level cache, 1.6 TB/s of memory bandwidth and AVX 10.2 alongside AMX matrix acceleration.
At Hot Chips 2026, Intel detailed more about its Crescent Island AI accelerator, which uses the Xe3P architecture. The accelerator will use liquid-cooled chips and HBM4 memory to serve inference workloads in data centers.
128 chips, 1.7 exaFLOPS, and 27 TB of HBM give Altman and crew a leg up over Blackwell, and maybe even Rubin
The newest member of NVIDIA's AI hardware family, at Hot Chips 2026 NVIDIA is diving into the use of LPUs as part of Vera Rubin clusters. The specialized chips from acquihire Groq are being tapped to offer significantly lower latency in the decode phase of inference
AMD showed off its MI400 Helios AI rack-scale infrastructure at Hot Chips 2026 going into some detail around how the rack operates
Dr. Semiconductor's latest project lights the path toward what he tells us he hopes becomes a cottage chip industry
At Hot Chips 2026, Samsung discussed how it plans to evolve the HBM base die to free up more package area for more efficient compute.
Nvidia has taken an undisclosed minority stake in Cloverleaf Infrastructure, a 2024-founded developer of land and power for AI datacenters. The Register reads the deal as the latest example of Nvidia using its war chest to patch cracks in the AI bubble: its newer GPUs need liquid cooling and power that most existing datacenters cannot supply, so it is funding the capacity that buys its chips.
Nvidia has rejected a report published by The Information that it plans to begin small-batch shipments of an LPU tailored for Chinese customers by the end of 2026.
Beijing wants most of each company's U.S.-licensed allowance, which the FT puts at up to 100,000 units apiece, kept off the mainland.
Google and its competitors have been pursuing custom chips to improve efficiency and reduce reliance on Nvidia.
Cerebras this week has introduced a major upgrade to its hardware ecosystem. The company is launching their first rack-scale AI inference system, the CS-4, which is powered by the upgraded WSE-3 Turbo processor.
Nvidia is trying to connect GPU customers with Nordic data-center operators as cheap power and available land fuel the region’s AI infrastructure boom.
China could become almost self-sufficient in high-end AI accelerators in 2026 as Chinese IHVs led by Huawei expected to supply 90% of AI processors used domestically.
Nvidia still has dominant market share in AI chips, but with competition picking up steam, the company is taking greater advantage of another key asset.
AMD says its 2025 rack-scale AI system is 4X more energy efficient compared to its 2024 AI solution, though does not produce actual benchmark results.
Intel’s Robert Hallock says he hopes enthusiasts “do the math” compared to AMD, highlighting that the company’s new core architecture will release in consumer processors before the data center.
The leading AI chipmaker has rebounded since late July.
PC Partner warns that graphics card prices will rise further in H2 2026 as memory costs climb and supplies tighten and entry-level boards will gain the most. This is how they gain money, says Jon Peddie.
No surprises on inflation and new financing developments in the artificial intelligence trade drove the record-breaking action on Wall Street last week.
We review the Minisforum N5 Max, a 64GB AMD Strix Halo system that combines 10GbE, a 5-bay NAS, and more into a single box The post Minisforum N5 Max Review with AMD Ryzen AI Max+ 395 appeared first on ServeTheHome.
This week, we tested AMD's BC-250 in gaming workloads, published an unredacted interview with an AMD executive, and published a slew of articles surrounding a new flashpoint in the ongoing AI buildout: optical interconnects.
Ukraine intelligence claims that Russia's latest S-71 'Monochrome' cruise missiles use Nvidia's Jetson Orin NX modules with AI capabilities, allegedly for terminal guidance.
Intel has seen a “sudden inrush” of demand for Raptor Lake CPUs, and it says they’ll remain a part of the company’s lineup for desktop builders “for years to come.”
We speak to Robert Hallock, Intel VP & GM of Enthusiast Channel Business, about Nova Lake rumors, how the company is focusing on DIY builders during RAMageddon, and how Raptor Lake refresh induced a paradigm shift for the company.
NeMo Switchyard brings GPT-5-style model routing to the mainstream
Technology giants have shelled out trillions to support insatiable AI demand and the infrastructure buildout.
Intel is raising $19.7 billion through a stock offering to strengthen its finances as it expands manufacturing capacity, develops next-generation process technologies, and is trying to attract major external foundry customers.
The 20-core SKU of the RTX Spark that we've known to exist for a long time scored 2,570 points in the single-core test and 23,126 points in the multi-core test. The second, 18-core cut-down SKU scored 2,541 points in the single-core test while its multi-core was 21,776 points.
We review the Minisforum N5 Max, a 64GB AMD Strix Halo system that combines 10GbE, a 5-bay NAS, and more into a single box The post Minisforum N5 Max Review with AMD Ryzen AI Max+ 395 appeared first on ServeTheHome.
Nvidia CEO Jensen Huang is pitching GPUs as long-term collateral to unlock $500 billion in funding. The question is how fast will his chips depreciate?
AMD's marketing director shared a screenshot of the Amazon CPU best sellers list, but the four-year-old, $80 Ryzen 5 5500 was at the top of the charts.
Swapping the executable of a game with a modified file will fool Steam into thinking it's opening that game when it's really just giving you unrestricted desktop access. This is against GeForce NOW's terms of service, though, so you run the risk of getting banned.
Nvidia CEO Jensen Huang argued that because its hardware is broadly adopted, flexible and transferable, lenders can underwrite compute as revenue-generating.
After recent news of price hikes on RTX 50-series GPUs in other regions, those same increases now appear to have come Stateside, as Newegg prices for some Blackwell cards have spiked as much as 39% compared to June 2026.
Intel’s orbital data center architecture uses powerful higher-orbit satellites to manage LEO constellations, reducing reliance on terrestrial control centers.
Colorful shows off the overclocking potential on new memory kits equipped with CXMT integrated circuits.
Anthropic has announced its building a team to co-design custom ASIC chips for AI inferencing workloads. This will give it greater control over its compute buildout and allow it to make its AI models specifically more efficient for its designs.
Services have recently popped up that will double your RTX 2080 Ti's memory to 22GB, but if you don't have a card to spare, you can now get a pre-modded 22 GB 2080 Ti for $499 from eBay.
AMD announced that it will acquire Taalas to build future model-specific AI inference chips that are purpose built to accelerate single models The post AMD to Acquire Taalas for Model Specific AI Inference Chips appeared first on ServeTheHome.
SoftBank on Thursday reported profit for its fiscal first quarter that beat market expectations.
SoftBank on Thursday reported profit for its fiscal first quarter that beat market expectations.
You can save $1,400 right now on this Omen Max gaming PC, fitted with an RTX 5090, 64GB RAM, and 2TB SSD storage, and on sale for $5,599.99.
Asian technology stocks rallied on Wednesday after Wall Street's record-setting session fueled optimism around artificial intelligence and growth stocks.
Elon Musk's SpaceX and xAI will exclusive use Nvidia AI accelerators for training and inference as companies believe Vera Rubin is the best AI compute architecture available today.
As cooling becomes a crucial element for economic efficiency of AI data centers, Frore claims that using is LiquidJet coldplate could increase efficiency of token generation by 15%.
SPONSORED POST: The AI factory concept encompasses the new AI stack, explain HPE’s Thierry Pienaar and NVIDIA’s Kaushik Shirhatti
Asian technology stocks rallied on Wednesday after Wall Street's record-setting session fueled optimism around artificial intelligence and growth stocks.
Anthropic is building a team for designing its own custom AI chips. The Claude maker said it would co-design hardware and models to help its technology run faster and more efficiently.
AMD reported record revenue in Q2 2026, including doubling its data center business year-over-year, but gaming revenue dived 31%.
The Lenovo ThinkPad X1 Carbon Gen 14 takes the long-running business notebook line into a new Intel Core Ultra generation. Our model 21V7006EUS pairs an Intel Core Ultra 7 355 with 32GB-class memory and integrated Intel graphics in a system The post Lenovo ThinkPad X1 Carbon Gen 14 Review appeared…
While the company's minimum OS specifications officially say 4GB, most PC builders know that 16GB is the bare minimum for a smooth experience on Windows 11. However, the memory chip shortage and the resurgence of 8GB laptops are likely forcing Microsoft to optimize Windows.
July 2026
It was a good but not great quarter as Apple worked to mitigate the memory shortage, which impacted the reported quarter and guidance.
This MSI Codex R2 gaming PC is on sale at Woot for $1,399.99 right now, delivering you a rig with a 10-core Intel Core i5-14400F, Nvidia GeForce RTX 5070, 32GB of DDR5 RAM, and a 1TB SSD.
We tested 20 motherboards for proper M.2 contact and were surprised at the results – not all make good contact.
This 27-inch LG OLED gaming monitor with a 280Hz refresh rate has dropped is on sale right now for $399.99 in a deal that'll save you $200.
High-bandwidth flash promises SSD-like capacities with HBM-like speeds, but it's not all unicorns and rainbows
Premium Nvidia GeForce RTX 5080 and RTX 5090 graphics cards are once again selling far above MSRP, with some Asus and MSI models climbing close to $5,000.
An Nvidia employee has been implicated in the Supermicro smuggling scandal, with his home and desk searched. He's been detained over allegations of forgery and breach of trust. Meanwhile, Nvidia is collapsing its buyer list in order to root out potential smuggling chains.
A GPU repair shop in the UAE is promising to upgrade the VRAM capacity of an RTX 2080 Ti from 11GB to 22GB by swapping the 1GB GDDR6 modules with 2GB chips and modifying the strap resistors to recognize the new VRAM.
We tested laptops with processors from Intel, AMD, Qualcomm, and Apple to see the differences in how they perform on and off the charger.
The MSI Cubi NUC AI+ 3MG is a compact, understated design, providing an Intel Core Ultra 9 386H in a tiny desktop. It’s easy to upgrade and relatively quiet but could do with a few more ports at the back and a more attractive price.
Etched, founded by three Harvard dropouts, has created new chips and memory components that speed up inference on any AI model -- no GPUs required, it says.
OpenAI CEO Sam Altman recently declared on the Relentless podcast that artificial intelligence has entered the technological singularity.
Taiwan's prosecutors say they've detained a man surnamed Chang on suspicion of falsifying business documents, after investigators searched his home and workplace.
Given the unprecedented surge in memory prices, we’re looking back on Intel’s LGA 1700 stack of CPUs to see how DDR4 and DDR5 match up with our modern gaming suite in 2026.
OpenAI is in advanced talks to lease SB Energy's 10 GW data center campus in Piketon, Ohio, with Nvidia in discussions to guarantee roughly $250 billion of the financing behind it.
With microscopic latencies and otherworldly write endurance, it was perfect for today's AI workloads - but gone before they arrived
The jockeying between Apple and Nvidia reflects a debate about the future of AI that has preoccupied investors this summer.
Apple is announcing an iPhone lease option weeks after raising iPad and Mac prices due to the global memory shortage.
Industry leading tech companies have formed an "Open Secure AI Alliance" that will build open-source models, agent harnesses, and cybersecurity tools, arguing that defenders need locally controlled AI after closed-model safeguards reportedly obstructed analysis of the OpenAI–Hugging Face breach.
Nvidia had held the top spot as the most valuable company since June 2025, when it took the crown from Microsoft.
At Computex 2026 Senao was showing off its latest wired SASE gateway, the SA9832v2, which is powered by Intel's elusive Atom X7 "Amston Lake" SoC The post This Senao SA9832v2 is an Intel Amston Lake-Powered Cloud SASE Gateway appeared first on ServeTheHome.
A Redditor is giving away an RTX 5060 gaming PC that's covered by RAM on the outside and has 32GB of RAM on the inside. You can join the sweepstakes simply by leaving a comment on the post.
Thermaltake has a no-brainer deal up on Woot right now where it'll let you have an $1,800 PC for as little as 1,177. For that price, you're getting an RTX 5060 Ti (8GB), a Core Ultra 7 265KF, 32GB of DDR5-6000 RAM, a 1TB M. 2 NVMe SSD, and a nice ARGB case.
Nvidia and SK Group enter $500 billion strategic partnership focused on long-term memory supply, 2 GW AI data center, and future AI infrastructure
Geekbench 7 is out with major overhauls to scoring, CPU and GPU benchmarks. We have already been testing with it The post Geekbench 7 is Out with a Major Overhaul appeared first on ServeTheHome.
Countering Intel’s recent moves, AMD is bringing its Strix Halo APUs to the realm of robots, and physical AI. Designed for 24/7 operation and a 10-year embedded lifecycle, X100 will also be offered as a Kria System on Module (SOM) robot developer platform.
Chinese open-weight models are gaining steam against leading offerings from American companies. OpenAI and Anthropic did not sign the letter.
Nvidia took the first stab with its Vera results earlier this week, and now AMD is biting back with SPEC results for its Zen 6 ‘Venice’ CPUs, as well.
When AMD's Helios meets giant wafers from Cerebras, it is not like when Odysseus meets with the Laestrygonian Giants, they collaborate to build an ultimate data center solution.
Here are five key things investors need to know to start the trading day.
Intel's revenue jumped 25%, the most robust growth for any period since the third quarter of 2011.
We go into a framework for normalizing NVIDIA Vera benchmarks from its latest whitepaper to some of AMD EPYC Turin's other parts The post Normalizing NVIDIA Vera Benchmarks to AMD EPYC Turin A Framework appeared first on ServeTheHome.
Google is developing a server chip, informally dubbed "Frozen v2," that would etch part of its Gemini model's architecture directly into the silicon.
IBM is now looking to improve productivity with artificial intelligence, including with a new coding tool called Bob.
AMD showed off its MI455X accelerator at its Advancing AI 2026 event, demonstrating its strong competitive performance, large HBM memory capacity, and Helios rack-scale architecture.
After over a year of teases, AMD has finally provided details on its 256-core Venice CPU with the Zen 6 architecture, now known as the Epyc 9996.
GM CFO Paul Jacobson says that the company's costs are expected to increase by $1. 5 to $2 billion, primarily due to increasing memory chip costs. The move comes as the RAM shortage is affecting the automotive industry, right as cars are requiring more memory and storage due to infotainment and ADAS features.
Chinese AI developer Z.ai (formerly Zhipu) has finished building a 1GW data center stocked exclusively with domestically made chips and has switched part of it on.
Intel rallied after the company reported blowout earnings and guidance.
The report comes at a critical time for optimistic investors like us at the Club.
TSMC reportedly intends to increase prices of wafers it processes citing demand, rising costs, and increased investments in new capacity.
Nvidia CEO Jensen Huang raised several points against the rising sentiment in Washington that U. S. firms should be prevented from accessing Chinese AI models.
Anthropic's agreement with AMD is the latest in a string of infrastructure deals the company has announced this year to increase its computing capacity.
Nvidia has just created an antidote to the virus that is AI misinformation. The company's new Synthetic Video Detector can help broadcasters assess 1080p footage at scale with processing times of just 22ms and up 92% accuracy for uncompressed videos.
NVIDIA this morning has released a trove of new technical details on Vera, their upcoming server CPU, as well as the Olympus CPU core. The company is also publishing the first SPEC CPU 2026 benchmarks, giving us our best look yet at the performance of the chip The post Diving Deeper on…
Nvidia gave Tom’s Hardware an exclusive look inside its previously undisclosed Engineering SuperLab near Nvidia HQ, where we saw Vera Rubin NVL72 in action.
Nvidia has detailed new features of its Rubin architecture.
As Nvidia continues to roll out Vera, its first custom CPU for agentic AI, it revealed that its last-gen Grace design has seen mass deployments, even as a standalone CPU for non-agentic workloads.
Fortinet is the latest Intel Foundry customer as the companies disclosed that the FortiSP6 will be a collaboration with Intel The post Intel Foundry Nabs A Custom ASIC Win with Fortinet appeared first on ServeTheHome.
Major benchmarks measure what AI can do. None measure whether it does what you mean: the distance between what you ask an AI to do and the unspoken assumptions about how you want the AI to do it. We propose a new metric: the Genie coefficient.
We have been seeing the AMD Instinct MI350P 144GB HBM3E PCIe accelerator everywhere over the past few weeks as this appears to be a popular GPU The post The AMD Instinct MI350P is a HBM PCIe AI Accelerator That Has Been All Over appeared first on ServeTheHome.
CuspAI is part of a new wave of companies looking to use AI to solve challenges in the physical world.
AMD's EPYC Venice is coming, and OEMs are eager to show off their upcoming wares. At Computex we caught MSI's CD182-S6091-X2 (DLC), a liquid cooled dual socket 1OU2N server node The post MSI Slyly Shows off an Upcoming DLC AMD EPYC Venice Platform With CD182-S6091-X2 Servers and Racks appeared first…
The Ryzen 7 7700X3D has suddenly become a solid value thanks to a $50 promo code, knocking its price down from $329 to just $279 on Newegg.
The high price for the garment is a sign that collectors are looking to bid on artifacts and collectibles from the artificial intelligence boom.
Save over 31% on the price of this Newegg 7700X3D bundle. CPU, motherboard, and RAM combo is just $450 after discounts.
Chief executive of Nvidia says the company is on track to produce 'giant amounts' of Vera Rubin-based machines, but fails to address rumored delays of Kyber NVL144 racks from 2027 to 2028.
We have been seeing the AMD Instinct MI350P 144GB HBM3E PCIe accelerator everywhere over the past few weeks as this appears to be a popular GPU The post The AMD Instinct MI350P is a HBM PCIe AI Accelerator That Has Been All Over appeared first on ServeTheHome.
The high price for the garment is a sign that collectors are looking to bid on artifacts and collectibles from the artificial intelligence boom.
TSMC announced a second-quarter profit, following its June revenue figures released earlier this week.
AMD's newest CPU, the Ryzen 7 7700X3D, costs $329 and is available exclusively at Newegg in Canada and the United States till the end of Q3 2026. It's a great gaming performer but there are better options if you want a good all-rounder chip for professional tasks as well.
Nvidia today announced that it's working with Japan's Noetra Corp. to build a 140-megawatt AI factory packing 27,500 Rubin GPUs and 13,750 Vera CPUs.
The 7700X3D is a 7800X3D with lower boost clock speeds, but it doesn’t deliver the same value as we’ve seen with previous versions of this segmentation.
TSMC will invest an additional $100 billion in the U. S. to build at least four more chipmaking plants and advanced packaging facilities in Arizona.
Beware of fab makers bearing press releases
Fireworks once relied heavily on revenue from coding startup Cursor, but has diversified in the past year as more companies reach for lower-cost AI models.
ASRock has entered the cooling market with the Phantom Gaming 360 LCD and Steel Legend 360 LCD AIOs. We’ve tested both liquid coolers with AMD’s Ryzen 9 9950X3D to benchmark their thermal proficiency.
Google has reportedly chosen Intel's EMIB-T over TSMC's CoWoS-L for its next-generation TPU, codenamed Humufish. But will Google be alone in its alleged decision?
We take a quick look at the Noctua NH-U14S DX4677 with the new Intel Xeon 600 series for workstations to see how it works The post Noctua NH-U14S DX4677 Quick Look With the Intel Xeon 600 Series appeared first on ServeTheHome.
The United States has licensed Chinese telecom giant ZTE to purchase restricted Nvidia H200 AI chips, but Chinese regulators and domestic procurement initiatives may limit the material impact of the change.
Experimental options discovered in AMD's latest Radeon driver suggest the company is preparing next-generation FSR technologies, but there's no confirmation whether 8x Multi-Frame Generation mode will ever ship in its current form.
Japanese chipmaker Rapidus discloses one more aspect of its strategy: to offer lower quotes than TSMC.
Pulls AI-powered image tweaker after allowing free-for-all
The Samsung PM1763 is a PCIe Gen6 NVMe SSD that the company said has entered production to support new AI servers The post Samsung PM1763 PCIe Gen6 Enterprise SSD in Production appeared first on ServeTheHome.
“The lowest-cost place to put AI will be in space, and that will be true within two years, maybe three at the latest,” SpaceX founder Elon Musk told the World Economic Forum in Davos this past January, as his company was preparing to go public. Later that month, SpaceX filed an application with the Federal Communications Commission for an orbital data center constellation of up to 1 million satellites in low Earth orbit, 500 to 2,000 kilometers above Earth. And just three days before the IPO, he discussed some initial design specifications for a new AI-1 satellite data center in a video interview.
All eyes on the Advancing AI 2026 event.
ASRock Rack has developed an unlikely edge server based on NVIDIA's industrial SoC, Thor. The Blackwell-era chip is being used to power the 2UXGI-THOR, a server aimed at the industrial and medical markets The post ASRock Rack Built an Edge Server Based on NVIDIA’s Thor Industrial SoC appeared…
AMD's Radeon RX 9070 GRE has received its first price cut since launching outside China, making the 1440p-focused RDNA 4 graphics card a more compelling alternative to Nvidia's RTX 5060 Ti 16GB.
SK Hynix CEO Kwak Noh-jung says the memory shortage will get even worse in 2027, and claiming the RAM crunch will last at least until the turn of the decade.
Apple sued OpenAI, including its own former employees, over the theft of trade secrets as both companies build up AI hardware businesses.
Intel's forgotten 14-core SKU from last year has received a sizable discount on Amazon, making it one of the best value propositions in CPUs right now. It performs amicably in gaming and professional tasks thanks to its 6 P-cores and 8 E-cores that can clock up to 5. 2 GHz.
SK hynix raised $26.5 billion in a record-breaking Nasdaq IPO, as it plans to channel the windfall from surging AI demand and sold-out HBM supply to fund new fabs.
Micron has said it will invest up to $3 billion in the US semiconductor supply chain, with $500 million of that going to GlobalWafers.
AMD's quiet revival of older Ryzen processors continues, with the Ryzen 7 4700LE now appearing in a prebuilt gaming desktop priced at $799.99.
Intel readies Xeon 'Dunlow' platform with 28 cores in LGA1954 packaging for entry-level servers and workstations.
A new 10-core engineering sample from AMD has surfaced on Geekbench, being identified as part of the Medusa Point family. It's likely the Ryzen AI 9 565 and its scores easily beat the Ryzen AI 9 HX 370 and even the Ryzen AI Max+ 395 in certain benchmarks.
Broadcom accuses Allstate of dodging VMware audits.
Micron announced a new round of investments aimed at boosting the U. S. semiconductor supply chain.
Meta is upgrading its Muse Spark artificial intelligence model under the leadership of AI chief Alexandr Wang.
The ₩1,350 trillion total combines a $520 billion semiconductor program with AI data center and robotics spending, mostly made up of corporate capex.
Nvidia lifts the veil a little bit more on its Vera CPU and reveals a single-thread performance monster — company claims a 1. 8x uplift versus x86 competition in agentic workloads and 1. 5x in coding.
Total graphics power figures for Nvidia's unanounced, unreleased RTX 50 Super-series graphics cards have appeared in Seasonic's PSU capacity calculator, revealing potentially higher TGPs of those products.
America's AI supply chain now starts and ends in the U. S. , while its most valuable middle steps remain entirely offshore until at least 2028.
The Samsung PM1763 is a PCIe Gen6 NVMe SSD that the company said has entered production to support new AI servers The post Samsung PM1763 PCIe Gen6 Enterprise SSD in Production appeared first on ServeTheHome.
Chinese memory and storage manufacturer Longsys expects to post a massive increase in profits due to the AI-driven chip shortage.
Brokerage consensus puts Samsung's full-year 2026 operating profit near 300 trillion won.
Probably an extension of long-running iThing innards arrangement, not a new custom AI accelerator deal
It marks Amazon's latest debt raise as it looks to buttress its massive investments in artificial intelligence.
The reported delay adds to concerns that Nvidia's breakneck annual release cadence is colliding with manufacturing limits.
The Ryzen AI Halo is a turn-key AMD local AI box that’s backed up with first-party software support, handy utilities, and plenty of documentation for local AI explorers. But the performance and application compatibility of the Ryzen AI Max+ 395 (aka Strix Halo) SoC still trails Nvidia’s GB10, and…
The reported delay adds to concerns that Nvidia's breakneck annual release cadence is colliding with manufacturing limits.
Huawei is reportedly preparing to enter South Korea's AI accelerator market with its Ascend 950 chips and Atlas 950 SuperPod, challenging Nvidia through aggressive pricing, amid a broader push to expand its AI ecosystem beyond China
A Redditor experience the deadly "Red Line of Death" on their Steam Machine, indicating GPU failure. They've just been playing for five minutes and then updated the console before experiencing the issue.
Intel confirms price increases for Core Ultra 200S Plus, Xeon 6 processors, cites market dynamics, rising costs, soaring demand.
Intel reportedly solves one of the key issues that plagued its 18A process technology, but others may still be there.
SummaryRFIC design is a complex “dark art” that limits progress in wireless technologies like 5G, autonomous vehicles, and satellite communications. Princeton researchers use reinforcement learning and inverse design to rapidly create RFICs from scratch. Diffusion models rapidly generate novel or human-interpretable RF layouts, achieving record performance and drastically reducing design time.
SK hynix announces major plan to spend $712.5 billion in its operations in South Korea, but the only detailed investments are spendings on a new NAND fab and a packaging facility.
Nvidia has announced a new business model under which it’ll be able to double-dip for revenue on the same silicon.
Four individuals suspected of smuggling Nvidia AI GPUs into China by using Singapore as a transshipment point are facing multiple charges. Singaporean authorities say they are not obliged to enforce foreign export controls but expects businesses in its borders to abide by them.
The day before President Donald Trump walked back his tariff policy turned out to be one of his busiest days of 2025 for stock buying.
AI server gold rush leaves budget laptops starving for chips as sub-$500 machines down nearly a fifth stateside
“The lowest-cost place to put AI will be in space, and that will be true within two years, maybe three at the latest,” SpaceX founder Elon Musk told the World Economic Forum in Davos this past January, as his company was preparing to go public. Later that month, SpaceX filed an application with the Federal Communications Commission for an orbital data center constellation of up to 1 million satellites in low Earth orbit, 500 to 2,000 kilometers above Earth. And just three days before the IPO, he discussed some initial design specifications for a new AI-1 satellite data center in a video interview.
June 2026
President Lee unveiled an 800 trillion won ($520B) public-private plan for four new Samsung and SK Hynix fabs, dwarfing the US CHIPS Act tenfold.
AMD confirms plans to incorporate low-power CPU cores into next-generation heterogeneous CPUs to lower power consumption and improve energy efficiency.
Micron CEO Sanjay Mehrotra said years of aggressive pricing pressure left the memory industry underinvested, contributing to today's supply shortages.
Anthropic will start an internal drug discovery program, part of a new push to sell artificial intelligence tools to drugmakers.
AI server gold rush leaves budget laptops starving for chips as sub-$500 machines down nearly a fifth stateside
SoftBank Group plunged, leading a broad sell-off in Asian technology stocks amid mounting concerns over the rising cost of AI infrastructure.
Onsemi and Synaptics to merge in a bid to build comprehensive platforms for robotics, physical AI applications.
We tested both CPUs across gaming, rendering, encoding, efficiency, and pricing to see if the Ryzen 7 5800X3D can keep up with the newer Core i7-14700K with DDR4.
PARTNER CONTENT: Shifting from legacy "connectivity pipes" to intent-driven intelligent services, the 6G-ready architecture introduces an Agent Service Network designed for the era of AI and human collaboration
Find the very best PC hardware deals during Amazon Prime Day.
AMD has re-released the Ryzen 7 5800X3D to provide some relief from high DDR5 prices, so we’re re-reviewing the CPU to see how it stacks up to current options around the same price.
We are at Qualcomm Investor Day 2026 where the company is talking High Bandwidth Compute, its HBM alternative, Its Dragonfly C1000 CPU, and AI accelerators The post Qualcomm Investor Day 2026 Data Center Announcements CPUs, AI Accelerators, and More appeared first on ServeTheHome.
The news involving OpenAI, the creator of ChatGPT, comes as Broadcom stock fights to get back on track.
ON Semiconductor said the deal bumps up its total addressable market by $30 billion, to $243 billion by 2030.
Artificial intelligence is the transformative, strategic technology of the early 21st century. It is significantly reshaping practically every aspect of our lives, including in ways that probably no one anticipated. Its rate of adoption and impact have been unprecedented when compared with other technologies.
TSMC has reportedly told customers to prepare for 5% to 10% price hikes across advanced chip nodes, extending beyond 3nm to include 7nm and some legacy processes.
Mini PC specialist Geekom has stuffed its Amazon webstore with a multitude of diminutive computers with discounts as deep as 34% off.
Chinese companies are paying as much as $82,000 for servers built around Nvidia's five-year-old A100 accelerator
Score 32GB DDR5 RAM from Corsair for as low as $214 in these two Newegg RAM combo deals with Gigabyte motherboards right now.
We are at Qualcomm Investor Day 2026 where the company is talking High Bandwidth Compute, its HBM alternative, Its Dragonfly C1000 CPU, and AI accelerators The post Qualcomm Investor Day 2026 Data Center Announcements CPUs, AI Accelerators, and More appeared first on ServeTheHome.
Micron said in its earnings report that its gross margin jumped to 84.9% from 39% a year ago.