The Wire · Chips
The silicon beat — GPUs, custom accelerators, HBM, advanced packaging, and the foundries and supply chains that gate every AI buildout.
OpenAI cut GPT-5.6 Luna API prices by 80% to $0.20 per million input tokens and $1.20 per million output, while Terra fell 20% to $2 and $12. Fast mode gives Sol up to 2.5 times Standard speed at twice the price. OpenAI says Sol-assisted kernel work lowered serving cost by 20% and improved token-generation efficiency by more than 15%, while Luna delivers year-old frontier performance at roughly six cents per task-dollar and nearly nine times the speed. The edition connects cheaper models to Amazon's reported $1.8m, 860%-over-budget coding task, Gemini Robotics 2 whole-body control, Nscale's Anyscale acquisition and Okta's roughly $200m Permiso deal.
303 points, 198 comments on HN
Read full story →Building a PC is a great experience. We sit there with a spreadsheet and multiple browser tabs, pouring over the best CPUs, the best GPUs , PC cases , motherboards , RAM deals, and many more components. The package arrives, and we are excitedly tearing into the boxes, putting the parts on the table, and we are ready to build! We reach into the toolkit and dig out the rusty old screwdriver we used to open a paint can back in 2014. We then spill standoffs and screws over the floor, ready for an unsuspecting foot at 2am! But wait, it doesn't have to be like this! We've all done this, made the mistakes, and now we share my knowledge and that of the Tom's Hardware staff so you don't have to hunt for the Amazon Prime Big Deals . So tool up, and let battle commence! Check out all the PC tool deals on Amazon Precision Kits All the gear, and lots of ideas? You need to start somewhere, and these kits have pretty much every tool that you could need to maintain a PC, all in a handy case. In my office/workshop, I have a general set of screwdrivers for general tasks, and a set of Hoto screwdrivers and bits for servicing my PC and any other parts I review, like motherboards, cases, fans, and more. NEX O1 PRO 3.6V Screwdriver Set: was $49.99 now $28.48 HOTO's wireless screwdriver has three torque settings, charges over USB-C, and looks surprisingly good. This kit comes with a sturdy box and all the bits you will need to keep your PC running. View Deal HOTO SNAPBLOQ S-A01 3.7V Electric Screwdriver: was $99.99 now $79.99 This HOTO SnapBloq 3.7V electric screwdriver includes 56 bits including Slotted, Phillips, Hexagon, Hexagon Security, Torx, Torx Security, Pentalobe, Standoff, Y-Shaped, Tri-Wing, Triangle, Square, U-Shaped, JIS, SIM Card Needle. It also has three torque settings, an integrated light, and charges via USB-C. View Deal 155 Piece Electronics Toolkit: was $59.99 now $37.9 This 155-piece kit has everything you need to get started with your first/next PC build. You get a p
Read full story →In its first live event in two years, Microsoft today announced its newest Surface laptop, outlined a host of changes coming to Windows 11, and shared a vision of how local-AI and agentic workflows could reshape personal computing for those in the dev community as well as home enthusiasts. Surface Laptop Ultra Originally teased in May 2026 , the new Surface Laptop Ultra will be Microsoft's first offering to take advantage of the new Nvidia RTX Spark SoC (system on a chip). The device has a starting price of $2,599, and comes in various SoC and memory configurations, with either a 5120-core or 6144-core Blackwell GPU and up to 128GB of LPDDR5x unified memory. It will begin shipping on October 16 and is available for pre-order now. While the new Ultra is clearly focused on the developer crowd, graphical performance was highlighted during the presentation to show off just how capable the laptop can be. By leaning on unified memory allocation rather than the physical VRAM limits of a traditional GPU, the laptop can assign its resources to handle modern gaming tasks as well as local AI development and deployment. The new Gears of War: E-Day was shown running on the system, meaning that even without a traditional gaming GPU, AAA titles are on the table. Read full article Comments
Read full story →AI may be taking our jobs, our RAM, and our storage... but there's always a light at the end of the tunnel. Amazon's Prime Day invariably triggers a rash of discounts at other retailers, and we just found the best SSD deal you'll come across until Black Friday at the earliest. The Teamgroup T-Force G50 Evo is a 4 TB solid-state drive with a PCIe 4.0 NVMe interface that can be yours for only $399.99 at Newegg with promo code FTTF9875 — or around 10 cents a gigabyte — with free shipping, to boot. Check out the Teamgroup T-Force G50 Evo 4 TB NVMe PCIe 4.0 SSD at Newegg. It's reasonable to wonder if 4 TB isn't too much of a drive to buy. It's likewise reasonable to rebut by stating that there's no end in sight to the AI shortages, that prices for NAND and hard drives have been climbing at insane rates, and that various modern games take up 150 GB to 200 GB of space. In other words: the wise course of action is to acquire this while the price tag only contains three digits before the dot. All-time low price The TeamGroup G50 EVO 4 TB at all-time low pricing. This drive can push 5000 MB/s in reads and 4,500 MB/s in writes, and its large capacity is a perfect fit for large game libraries. View Deal This unit ought to be capable of pushing 5,000 MB/s in sequential reads and 4,500 MB/s in writes. The endurance rating is quite generous for a consumer drive, at 1,200 TB written, and the quoted MTBF is 1,500,000 hours. These figures mean the drive is likely to last longer than the rest of your computer. Teamgroup backs up the figures above with a three-year warranty, and there's a graphene heatsink resting atop the chips to keep temperatures low and NAND happy. You can snag the $399 Teamgroup T-Force G50 Evo 4 TB NVMe PCIe 4.0 SSD at Newegg.
Nvidia's RTX Spark is arriving soon, with devices shipping on Friday, October 16. Pre-orders for RTX Spark from brands like Asus, Microsoft, Lenovo, HP, and Dell are live now, including the new Surface Laptop Ultra . Additional devices will be available post-launch, including small form factor desktops. Further below, you'll find the full list of all available models in detail, but we also have more general links here: Pre-order RTX Spark devices at Best Buy Pre-order RTX Spark devices at B&H Pre-order Surface Laptop Ultra at Microsoft Best Buy has a limited lineup of devices to start, ranging from the Asus ProArt P14 with 24GB of memory for $2,599 up to the Surface Laptop Ultra with 48GB of memory for $3,999. If you want to scale up the RTX Spark's maximum 128GB of memory, you'll need to shop elsewhere. Nvidia RTX Spark 128GB and 64GB laptop pre-orders ProArt P16 RTX Spark: $4399.99 The Asus ProArt P16 is a 16-inch laptop with the RTX Spark chip, which includes the 20-core Grace CPU and a Blackwell GPU, along with up to 128GB of unified memory. View Deal Asus has the ProArt P16 available with either 64GB or 128GB of unified memory directly from its web store, starting at $4,499.99 for the 64GB configuration. If you upgrade to 128GB of unified memory, the laptop comes in at $6,999.99 with 2TB of storage. The HP OmniBook X 14 is a 14-inch laptop with the RTX Spark chip. It includes a 3K OLED touch display at a 120Hz refresh rate, along with up to 128GB of unified memory. View Deal The HP OmniBook Ultra 16 is a 16-inch laptop with the RTX Spark chip and supports up to 128GB of unified memory. View Deal HP has two models available directly in its web store, both of which support up to 128GB of unified memory (along with 64GB, 48GB, 32GB, and 24GB). Both models are significantly cheaper than the ProArt P16 with 128GB of unified memory. The OmniBook X comes in at $5,109, while the OmniBook Ultra runs $5,309, both with 1TB of storage. The two models are very similar, diff
If you thought Windows Recall was creepy, buckle up. Microsoft is about to shoehorn even more AI into its desktop operating system, giving Copilot access to every file on your PC. But don't worry, Redmond insists it's totally secure and safe since the agents powering the functionality will run entirely on your PC — some of the time. Microsoft describes the new functionality, announced alongside a new Nvidia N1X powered Surface Laptop, as "hybrid intelligence." Built atop Microsoft's HydraFusion model router, Windows apps and Copilot features will soon be able to use local models when it makes sense and cloud-based models when it doesn't. For example, Windows might use a local model for simple or privacy-sensitive tasks, or decide complex or performance-sensitive requests need OpenAI's GPT-6 or Anthropic's Claude. Local agents will run in Microsoft's Execution Containers (MXCs), agent sandboxes with fine-grained permissions applied at runtime. According to Microsoft EVP of Windows and Devices Pavan Davuluri, MXCs mean Windows is now "the most secure platform for agents." (That claim will surely age well because models never escape their sandboxes and routers never misbehave.) To support this hybrid intelligence functionality, Microsoft is extending support for Windows ML, its inference framework built on the ONNX runtime, to Llama.cpp – one of the highest-performing inference engines for local AI inference. By adding support for Windows ML, Llama.cpp should be able to take advantage of a broader range of accelerators. Microsoft is rolling out its hybrid intelligence functionality to GitHub Copilot — its AI code assistant — next week and its Copilot app over the next few months. The new Copilot features look and sound suspiciously like OpenClaw in a business suit. Copilot's agentic functionality will not have access to your entire file system, opening the door to advanced search functionality and productivity tasks through integrations with Microsoft Office. In a demo
Microsoft revealed the specs and price for its Surface Laptop Ultra, AI PCs that run on Nvidia chips that are designed to run AI models and agents.
Look out, Intel! Microsoft released its first Surface notebook powered by Nvidia's RTX Spark system-on-chip (SoC) during an event in San Francisco on Wednesday. The Surface Laptop Ultra is one of several new Copilot+ PCs powered by the N1X chip that Microsoft says is aimed at "builders," which are available for preorder starting today and expected to begin shipping next week. The machine is arguably the vehicle for a slew of new AI features including hybrid compute functionality that dynamically routes workloads to on-device or cloud models depending on factors like connectivity, privacy, cost, and complexity. The aluminum clad system largely keeps with Microsoft's existing design ethos. It features a high-res, 15-inch mini-LED display capable of reaching a peak brightness of 2,000 nits in Microsoft's signature 3:2 aspect ratio, a standard ten-keyless keyboard layout, haptic trackpad, and a slew of ports including one you won't find on a MacBook Pro. Connectivity includes three USB-C ports, including one equipped with magnets for MagSafe-like connectivity, full-size HDMI, an SD card reader, and a solitary USB-A port because after more than a decade still can't shake the aging connector. But arguably the notebook's most interesting feature is the inclusion of Nvidia's RTX Spark N1X SoC. As we reported earlier this year, these aren't exactly new chips. The N1X is based heavily on Nvidia's now year-old GB10 platform, with a few CPU and GPU cores fused off on lower end models to meet pricing. This tells us that the Surface Laptop Ultra should ship with either an 18- or 20-core CPU designed by MediaTek and Arm and an Nvidia Blackwell-based RTX GPU with between 5,120 and 6,144 cores. In terms of memory, Microsoft says the notebook can be equipped with up to 128 GB of memory, enough to serve large language models up to 284 billion parameters in size, if you can live with just 1.6 bits of precision. As we've previously covered, the GPU is a pretty potent one with a peak the
The first array of laptops powered by Nvidia's new RTX Spark chip are designed to compete with high-end MacBook Pros - and they've got some high-end prices to match. The flagship Surface Laptop Ultra and RTX Spark laptops from Dell, Lenovo, MSI, HP, and Asus start as low as $2,599 for configurations with an 18-core […]
One of Donald Trump’s biggest goals as president is to increase domestic manufacturing and reshore supply chains so that the most in-demand technologies in the world can be 100 percent made in the United States. By executive order last December, he launched his “Genesis Mission,” an ambitious plan seeking to leverage US dominance in AI to maintain or seize an edge in advanced robotics, biotechnology, semiconductors, and nuclear technologies. According to Trump, that plan works best with self-sustaining supply chains. Not even neighboring countries or longtime allies should be relied upon to manufacture a single component of any US product in his ideal world. The Consumer Technology Association (CTA) has long warned that Trump’s reshoring goals are not grounded in reality, but only recently has America’s largest technology trade association attempted to estimate what it would cost businesses and consumers to follow through on Trump’s plan. Read full article Comments
Training the largest AI models is no longer simply a question of adding more GPUs. As clusters grow, power availability is becoming a hard physical constraint, forcing infrastructure teams to consider how a single training workload can operate across multiple data centres. That creates a very different networking problem. Traditional datacenter interconnect was designed to move traffic between sites. AI training demands something more exacting: huge, synchronous flows, minimal packet loss and tightly coordinated communication between GPUs. If one part of the cluster stalls, the impact can ripple across the entire job. In this interview, The Register’s Tim Phillips speaks with Rakesh Chopra, SVP, Silicon and Systems Architecture, Cisco Fellow, about what Cisco calls the “Scale-Across” imperative and why it believes distributed AI infrastructure requires a new approach to routing. Rakesh explains how the demands of AI are changing the role of the network, from simple transport to an integral part of the compute system. He discusses the challenge of making geographically separate facilities behave more like a single deterministic machine, and why buffering, high-speed coherent optics and tightly integrated silicon matter when workloads span long-distance fiber links. The conversation also looks inside the cluster itself. Rakesh outlines how Cisco’s Silicon One architecture and Intelligent Collective Networking are designed to manage synchronized bursts of GPU traffic, reduce bottlenecks and improve job completion time. Power efficiency is another central theme, including the trade-off between network consumption and the energy available to GPUs. Security and longevity also come into focus. The interview covers hardware-accelerated MACsec and IPsec, as well as the role of programmability in helping infrastructure adapt as AI workloads continue to evolve. For infrastructure and datacenter leaders planning for larger, more distributed AI environments, the discussion offer
There’s been something of a change in the world of motorsport over the past few years. It has to do with where the money comes from. In the old days, it was tobacco and, to a lesser extent, booze. Today such things are mostly banned, and when you look at a racecar covered in sponsor logos, you’re likely to see something crypto-, cyber-, or AI-related. This makes sense—an ever-larger share of our GDP is wrapped up in those segments. They’re the ones with all the money, and it takes money to go racing. So at the amateur level, we’ve seen tech CEOs replace dentists as the privateer of choice: This past weekend’s Petit Le Mans included co-founders of GitHub, CrowdStrike, Shopify, and the creator of Ruby on Rails, to name a few. But this is an instance where a sponsor can provide tangible benefits to their motorsports partner beyond a bank transfer, as Chip Ganassi Racing (CGR) and OpenAI have been showing in IndyCar this year. Read full article Comments
<p><strong><a href="https://mistral.ai/news/mistral-large-4/">Introducing Mistral Large 4: Le chonk</a></strong></p> Mistral are back in the game. Today they're releasing a preview of Mistral Large 4, a 1 trillion parameter, 49 billion active parameter model trained on their own cluster of 3,800 NVIDIA Grace Blackwell GPUs.</p> <p>The preview is available via their API. They promise to release the open weights model at the "end of this month".</p> <p>The model only supports two reasoning levels - "none" and "high" - via the Mistral API. Here are <a href="https://tools.simonwillison.net/markdown-svg-renderer?url=https%3A%2F%2Fgist.github.com%2Fsimonw%2F0db8b8196e57711b352f6c5857cf5b35">both pelicans</a> - the "high" one looks better, though surprisingly it only used 2,717 output tokens compared to "none" which used 3,275:</p> <p><img alt="It's good. The pouch is great, the bicycle frame is the right size, it has feet on pedals. Both pedals appear in front of the frame though. Nice gradients." src="https://static.simonwillison.net/static/2026-10-06/mistral-large-4-pelican.webp" /></p> <p>On Artificial Analysis <a href="https://artificialanalysis.ai/models/mistral-large-4">it scores 38</a>, just behind DeepSeek 4.1 Flash, which is a 552B model. It's a <em>huge</em> improvement on last December's Mistral Large 3, which drew <a href="https://tools.simonwillison.net/markdown-svg-renderer?url=https%3A%2F%2Fgist.github.com%2Fsimonw%2F0df5e656291d5a7a1bf012fabc9edc3f#response-1">this terrible pelican</a> and <a href="https://artificialanalysis.ai/models/mistral-large-3">scored 9 on AA</a>.</p> <p>It's certainly not a Fable-class model, but it's great to see Mistral put out a model that's back to being maybe about 6 months behind the frontier. <p><small></small>Via <a href="https://news.ycombinator.com/item?id=49977979">Hacker News</a></small></p> <p>Tags: <a href="https://simonwillison.net/tags/ai">ai</a>, <a href="https://simonwillison.net/tags/generative-ai">generative-ai<
We test the Gigabyte W775-V10-L01, including getting over 2.2B tokens/ day on the Blackwell Ultra GPU, 800Gbps from the ConnectX-8 SuperNIC, and testing the NVIDIA Grace CPU with the C2C link between it and the GPU The post Gigabyte W775-V10-L01 Hands-on Bringing NVIDIA GB300 Deskside appeared first on ServeTheHome .
<figure><div><img src="https://imgproxy.divecdn.com/tThVeTmTqwiND5hQDvss8fAGHBMdcLxGWFx_21c9l5k/g:ce/rs:fill:1600:900:1/Z3M6Ly9kaXZlc2l0ZS1zdG9yYWdlL2RpdmVpbWFnZS9HZXR0eUltYWdlcy0xNDU1ODY5Njk2LmpwZw==.webp"/></div></figure><p>The first round of awards under the Maryland Next Generation Energy Act fell short of the state’s 800-MW target as the PJM Interconnection’s capacity crunch continues.</p>
An independent briefing for builders: the whole field read continuously, every story scored for relevance, and the noise left off the page.
300+ curated sources. Every story scored 1–10 for builder relevance by Claude's frontier model. The filler never makes it to the page.
GPUs, datacenters, power deals, and inference economics: the infrastructure layer that decides what every builder pays. Our signature coverage.
Every story is sourced. Every score is computed. We show our work and link to originals.
Every briefing closes with The Call: one falsifiable claim with a date on it. When we're wrong, we say so in print. Opinions are cheap; ours get scored.