Skip to content
VibekollenBETAVibekollen

Source

NVIDIA

35 items in the feed. The texts are the sources' own descriptions — the content belongs to NVIDIA.

BlogNVIDIA

NVIDIA DGX Spark 64GB Gives Developers More Ways to Build and Scale Local AI

Local AI is becoming more useful by the token. As AI agents move from experiments into everyday development, increasingly capable open models are shrinking to fit on more devices, giving builders more to run locally. Coming this month, NVIDIA DGX Spark will be available with 64GB of unified memory from top manufacturer partners — Acer, ASUS, Dell, Gigabyte, HP and MSI — giving developers, researchers and AI enthusiasts a new configuration with DGX OS and the NVIDIA AI software stack ready to use from day one. The new SKU runs capable local agents on device — privately, without cloud dependency. And when workloads grow, two units can cluster together via NVIDIA Sync Cluster Assistant without any additional setup. A New Starting Point for Personal AI Supercomputing DGX Spark combines NVIDIA Grace Blackwell compute, unified memory, NVIDIA ConnectX-7 networking and an NVIDIA CUDA-accelerated AI software stack in one system. It’s a complete local AI platform for agents, inference, fine-tuning, data science and edge development. The compact, personal AI supercomputer provides a place to experiment with models and developers’ own data without turning to a cloud instance for every task.

2 Oct blogs.nvidia.com

BlogNVIDIA

How NVIDIA GPUs Help Accelerate OpenAI’s GPT-6 Astra Ultrafast

GPT-6 Astra Ultrafast, running on NVIDIA Blackwell GPUs, is available now in the OpenAI API and to eligible ChatGPT Work and Codex users. Accelerated by inference optimizations through OpenAI’s models that tap into the capabilities of the NVIDIA Blackwell architecture, Ultrafast offers up to 8x faster token generation than the Astra Standard mode. For developers, faster generation can shorten coding agents’ edit-test-debug cycles, reduce the time spent generating responses between tool calls and make interactive applications feel more responsive. A faster response matters most when it’s repeated across a workflow: an agent writes code, uses a tool, checks the result and decides what to do next. Ultrafast brings Astra’s capabilities into these time-sensitive loops. NVIDIA AI infrastructure helps OpenAI serve more useful model outputs when developers need it. “NVIDIA’s deep investment in tooling and documentation has enabled us to make our models exceptionally good at programming Blackwell and Rubin GPUs,” said Philippe Tillet, inference lead at OpenAI.

2 Oct blogs.nvidia.com

BlogNVIDIA

Productive, Durable, Fungible: How NVIDIA AI Factories Maximize Return on Investment

AI factories are built by the megawatt, even by the gigawatt. Each megawatt factory costs roughly $60 million, and AI factory operators will only commit capital on that scale with a clear view of the return on investment. Three key things shape AI factory returns: Earning capacity: What the factory could earn in a year if it sold every token it can produce. Useful life: How long its AI hardware keeps earning. Demand: How much demand there is for those tokens. Strength cannot fully offset weakness in another. High earning capacity counts for little if the factory sells only part of what it can produce. High demand matters little if it stops producing at full capacity in just a year. Nor are the three independent. A factory that can run more kinds of workloads finds more demand, keeping it earning year after year. NVIDIA AI factories are engineered to maximize all three. They’re: Productive: Delivering the highest throughput per megawatt and the lowest cost per token, which maximizes their earning capacity. Durable: NVIDIA GPUs and systems keep earning years after they ship, extending useful life.

1 Oct blogs.nvidia.com

BlogNVIDIA

NVIDIA Opens Applications for 2027–2028 Graduate Fellowships With Awards Up to $60,000

Bringing together the world’s brightest minds and the latest accelerated computing technology leads to powerful breakthroughs that help tackle some of the biggest research problems. To foster such innovation, the NVIDIA Graduate Fellowship Program provides grants, mentors and technical support to doctoral students doing outstanding research relevant to NVIDIA technologies. The program, in its 26th year, is now accepting applications worldwide. It focuses on supporting students working in AI, machine learning, autonomous vehicles, computer graphics, robotics, healthcare, high-performance computing and related fields. Awards are up to $60,000 per student. Since its start in 2002, the Graduate Fellowship Program has awarded over 220 grants worth around $8 million. Students must have completed at least their first year of Ph.D.-level studies at the time of application. The application deadline for the 2027-2028 academic year is October 30, 2026. An in-person internship at an NVIDIA research office preceding the fellowship year is mandatory; eligible candidates must be available for the internship in summer 2027. For more on eligibility and how to apply, visit the program website.

30 Sept blogs.nvidia.com

BlogNVIDIA

From Training to Production, NVIDIA and CoreWeave Close the Loop on Agentic AI

Building on nearly a decade of co-engineering, CoreWeave has built NVIDIA compute, networking and software into a cloud purpose-built for AI that’s still returning on investment across multiple generations of deployment. Now, CoreWeave is bringing the next generation of NVIDIA infrastructure to production. At CoreWeave Fully Connected, running this week in San Francisco, CoreWeave announced availability of NVIDIA Vera Rubin NVL72 systems with Spectrum-X 102.4T Ethernet networking. Cognition, the applied AI lab behind the Devin AI software engineer, is the first customer running production workloads on Vera Rubin. CoreWeave will also offer NVIDIA Vera, the first CPU built for AI agents. In addition, CoreWeave launched CoreWeave Forge, a connected environment for training, evaluating and improving models and agents on NVIDIA accelerated computing. “NVIDIA accelerated computing delivers value across generations,” said Ian Buck, vice president of hyperscale and high-performance computing at NVIDIA. “CoreWeave’s NVIDIA V100 GPUs are still running customer workloads nearly a decade after Volta launched, even as CoreWeave brings Vera Rubin NVL72 into production.

30 Sept blogs.nvidia.com

BlogNVIDIA

How Open Science Can Help Researchers Prepare for the Next Pandemic

When COVID-19 emerged, scientists had a crucial advantage: Decades of prior research on coronaviruses meant they understood the virus’ key proteins well enough to design vaccines in record time. The next pandemic may not offer the same head start. To help improve the odds, NVIDIA has joined a coalition of global research organizations, including Google DeepMind and the European Molecular Biology Laboratory’s European Bioinformatics Institute (EMBL-EBI), to release predicted 3D structures for the protein complexes of more than 2,800 viruses — openly available to any scientist, anywhere, through the AlphaFold Database. The structures in the newly released dataset were inferred using AlphaFold2 — Google DeepMind’s AI model for predicting how proteins fold into 3D shapes — with optimization from NVIDIA BioNeMo Inference Runtime. This allowed the team to scale inference to thousands of viral proteomes, predicting the complexes, or groups of interacting proteins, encoded within each virus. “Our ambition with the AlphaFold Database has always been to democratize access to foundational biology at scale,” said Risha Patel, life sciences partnerships manager at Google DeepMind.

24 Sept blogs.nvidia.com

BlogNVIDIA

Sakeena Fiza Helps NVIDIA Hardware Succeed at Scale

When Sakeena Fiza describes her work as a validation engineer at NVIDIA, she does so in terms more befitting a detective story than a world-class engineering lab. “Validation engineers look in the shadows and shine a light into every corner,” Fiza said. “Every time we get a system, our first thought is: how can it break?” And when it does? “I always like to think of it as a mystery to solve,” she said. At NVIDIA, the systems Fiza and her colleagues in the data center systems engineering lab investigate are the engines of the AI era. Her work begins before the rest of the world knows a product exists — in the lab — when a new system first receives power. Components are brought up one by one, boards are integrated, firmware and software teams swarm, and engineers watch for the first signs of life. One of Fiza’s earliest and most enduring memories of working at NVIDIA is the collective joy she experienced when she saw the NVIDIA Rubin GPU working for the first time at a system level. “It literally just said, ‘NVIDIA Corporation Device,’” she recalls.

23 Sept blogs.nvidia.com

BlogNVIDIA

At AI Day Singapore, NVIDIA and Partners Showcase AI Advancements Across Southeast Asia

NVIDIA AI Day Singapore, which takes place Sept. 22-23 at the Raffles City Convention Centre, is offering attendees opportunities to explore the hands-on training, expert-led sessions and advanced tools to accelerate their work in AI and high-performance computing. At the event, NVIDIA and its partners are showcasing breakthrough AI advancements across the Southeast Asia region at large. Read more about these announcements below. NVIDIA Accelerates Public Sector AI from Pilot to Production in Southeast Asia AI is becoming a matter of national strategy, with governments looking to move from pilots to production and deliver impact at scale, while building trusted AI capabilities that reflect local languages, cultures, priorities and economic needs. NVIDIA is working to enable all nations to be AI nations — providing the technology, infrastructure, ecosystem and expertise needed to make this possible. To accelerate this transition across Southeast Asia, NVIDIA is helping nations move AI from experimentation to production-scale deployment through open models, developer tools and a broad partner ecosystem.

23 Sept blogs.nvidia.com

BlogNVIDIA

NVIDIA Isaac ROS 5.0 Advances Agentic, Open Source Robotics Development

To build and deploy sophisticated robotics applications that can perceive, reason and act in dynamic environments, developers need new physical AI models and tools. The ROS open framework is a project from Open Robotics that helps humans build robots. NVIDIA Isaac ROS 5.0 — a collection of GPU-accelerated packages built on ROS, released today at the ROSCon conference in Toronto, Canada — helps humans and AI agents build robots together. The release introduces new agentic workflows and platform support to help developers build, customize and deploy robotics applications faster. ROS provides the open source foundation for much of modern robotics development, giving developers common tools, libraries and standards for building and connecting robot applications. NVIDIA Isaac ROS brings NVIDIA accelerated computing, physical AI models and production-ready libraries to the nearly 1.3 million ROS users, helping developers build high-performance robotics applications using free, familiar, open source tools.

22 Sept blogs.nvidia.com

BlogNVIDIA

NVIDIA Launches DSX Ready to Qualify Power and Cooling Products for AI Factories

Every AI factory needs power and cooling that fit its computing architecture. As AI infrastructure expands, power, cooling, water, site and grid constraints are shaping what builders can deploy. Choosing products that fit the complete factory design helps builders turn computing capacity into useful AI output. To help builders make those decisions, NVIDIA is introducing NVIDIA DSX Ready, a qualification program for partner products and solutions that meet applicable NVIDIA DSX AI factory reference design requirements. The program launches with two initial categories: battery energy storage systems (BESS) and cooling distribution units (CDUs). Category-specific requirements and review through the program help builders evaluate offerings with greater confidence, reduce integration risk and move toward deployment. Qualified Building Blocks for Building AI Factory The NVIDIA DSX AI factory platform unifies AI factory design and operations across compute, networking, power, cooling, facilities and software. It helps partners design and operate the factory as one system to produce more useful AI output within available power, cooling, water and grid constraints.

21 Sept blogs.nvidia.com

BlogNVIDIA

Why Deploying Physical AI at Scale Demands Safety at Every Layer

Physical AI is moving rapidly from research to large-scale deployment. By 2035, ABI Research projects an installed base of 49 million level 3-5 autonomous vehicles (AVs), while Omdia estimates that roughly 60 million industrial robots will be deployed between 2026 and 2035. As these machines enter roads, factories, warehouses and other environments shared with people, safety must scale with them. Physical AI safety means proving that AI-driven machines — AVs, humanoid robots, industrial robots and more — behave safely when their decisions turn into physical action. That requires safety across the hardware, software, AI, operating environment and deployment lifecycle — not a one-time check before deployment. Why Is Safety the Key to Scaling Physical AI? After years of testing and benchmarking, AVs continue to expand commercially. That progress has required developers to demonstrate how automated systems address potential hardware and software failures, limitations in intended functionality and AI-specific risks. Robotics is approaching a similar inflection point as autonomous machines move into factories, warehouses and other environments shared with people.

21 Sept blogs.nvidia.com

BlogNVIDIA

From Enablement to Execution, Egypt’s AI Ecosystem Reaches Production Scale

Today, Egypt’s AI builders gathered in the Grand Egyptian Museum for a reception that highlighted the nation’s rapidly growing AI ecosystem — spanning AI natives, developers, researchers, startups and enterprises — building applications across industries. The event included a keynote from Paolo Guglielmini, vice president of EMEA at NVIDIA. Ahmed Mostafa, regional AI adoption lead for the Middle East and Africa at NVIDIA, delivered a session on “Why Accelerating Every Layer Matters,” exploring NVIDIA’s full-stack approach to AI development and deployment. The event also featured a panel moderated by Basil Fateen, head of startups for the Middle East and Africa at NVIDIA, with participation from startups across smart spaces, healthcare, cybersecurity and robotics. In Egypt, the NVIDIA Deep Learning Institute learner base grew more than tenfold in a single year. In June, the National Telecommunications Regulatory Authority of Egypt licensed Hassan Allam Data Centers to build and operate data centers in the country. Under that license, Hassan Allam Utilities and investment firm A15 agreed to develop a new data center, an estimated $400 million investment.

21 Sept blogs.nvidia.com

BlogNVIDIA

AI Security Is an Engineering Problem — How to Solve It at Every Layer of the Agent Stack

AI security is an engineering problem. That means defined security requirements, enforceable controls, named owners and evidence that protections work. As AI becomes more capable, the industry must accelerate security engineering, broaden access to defensive tools and share what works faster. Technology Changes, Security Fundamentals Endure The internet and cloud computing changed how software operates, while core security responsibilities endured: establish identity, control access, limit exposure and verify that protections work. AI agents introduce new capabilities — reasoning, using tools and adapting actions based on the data they encounter. Those capabilities require applying established principles to new operating conditions. This pace creates pressure. Organizations want the productivity benefits of AI while the practices to govern and secure these systems are still developing. Security Depends on the Full Agent Stack Applications depend on code, data, identities, services and infrastructure. Security depends on how those components work together — and AI agents extend that system.

21 Sept blogs.nvidia.com

BlogNVIDIA

5 Companies Using NVIDIA AI for Clean Energy

NVIDIA showcases five companies using AI to accelerate the shift to clean energy. ThinkLabs AI uses digital twins to reduce the time to evaluate grid interconnection applications from 30–45 days to just two minutes. Atomic Canyon develops AI tools for the nuclear power industry, while Redwood Materials builds battery systems from recycled electric vehicles to power AI factories without waiting for grid expansion. TerraPower uses digital twins to speed up deployment of new nuclear plants, and Commonwealth Fusion Systems employs AI to accelerate fusion energy research.

21 Sept blogs.nvidia.com

BlogNVIDIA

Cute Critters Come to the Cloud: ‘Aniimo’ Launches on GeForce NOW

NVIDIA is launching eleven new games on its GeForce NOW cloud service this week, including the free game Aniimo from Pawprint Studio — an open-world adventure where you catch and collect creatures and can stream directly without downloading the game's 45GB. The James Bond game 007 First Light also receives an update with path tracing for more realistic lighting and shadows. Two other new games are Active Matter, a military shooter set in a fractured multiverse, and the space mystery Outer Wilds — all streamable on Steam Deck, Firefox and other devices.

17 Sept blogs.nvidia.com

BlogNVIDIA

NVIDIA Vera Rubin NVL72 Delivers Leading Performance in MLPerf Inference v6.1 Debut

NVIDIA presents its new Vera Rubin NVL72 system, which shows significant performance improvements in MLPerf Inference v6.1 tests. The system delivers up to 3.7x higher throughput than its predecessor GB300 NVL72 on the Qwen3-VL benchmark, and 2.5x higher on DeepSeek-R1. When four GB300 NVL72 racks were scaled up together, 99 percent scaling efficiency was achieved — throughput grew nearly linearly as more GPUs were added. NVIDIA also achieved performance gains through software optimization: version 6.1 was up to 1.6x faster than 6.0. For AI agents solving complex multi-step problems, Vera Rubin delivered 30x better performance than GB300 in testing.

16 Sept blogs.nvidia.com

BlogNVIDIA

Emerald AI, Google and NVIDIA Launch Alliance to Advance Flexible AI Data Centers

Emerald AI, Google, and NVIDIA have launched the AI Energy Management Alliance (AEMA), a coalition helping AI data centers flexibly adjust their electricity use based on grid conditions. Instead of always drawing the same amount of power, these centers can shift computing tasks, use storage, or adapt when the grid is strained. This allows more computing capacity from existing infrastructure, reduces environmental impact per watt, and gives utilities greater confidence to connect AI facilities faster. The alliance focuses on performance rather than specific technology, and sets clear rules for how data centers must behave during grid disturbances and emergencies.

16 Sept blogs.nvidia.com

BlogNVIDIA

University of Manchester Uses NVIDIA Earth-2 to Forecast Air Pollution Across the UK

Researchers at the University of Manchester are using NVIDIA Earth-2, an AI model originally developed for weather forecasting, to predict air pollution across the United Kingdom. Air pollution causes approximately 30,000 deaths annually in the UK, but traditional chemistry-based models are too slow and expensive for detailed forecasts. The team trained the Earth-2 CorrDiff model on Isambard-AI, the UK's national AI supercomputer, in just two days using a year's worth of air pollution data. The model can now run on a smaller desktop computer (DGX Spark) and be used to create future scenarios, alert healthcare providers about high pollution levels, or respond to events like wildfires.

16 Sept blogs.nvidia.com

BlogNVIDIA

‘Now We Can Know Everything and Do Anything,’ Jensen Huang Says at Dreamforce

At Salesforce Dreamforce, NVIDIA founder Jensen Huang unveiled Koa, Salesforce's first CRM reasoning model—an AI model trained on NVIDIA's Nemotron 3 Super. Koa was built by fine-tuning Nemotron on data from nearly three decades of real CRM usage and delivers 3x fewer errors than leading models on CRM tasks like updating opportunities or routing cases. Huang emphasized that safety is paramount but that development and rapid innovation need not come at the expense of security; Salesforce uses no customer data during training or operation, and the model runs entirely within Salesforce's own infrastructure.

16 Sept blogs.nvidia.com

BlogNVIDIA

From Megawatts to Tokens: How NVIDIA Maximizes AI Factory Production

NVIDIA introduces DSX, a system to optimize power consumption in AI factories (large data centers). A test in California showed that a power grid could send signals to a data center, which then automatically reduced power consumption from four to three megawatts without disrupting critical workloads. Lambda Labs, a cloud provider, found that the same power budget could produce 24 percent more work when managed intelligently. The system allows data centers to respond to grid needs by pausing less important computations while critical tasks continue.

15 Sept blogs.nvidia.com

BlogNVIDIA

AI Infra Summit: NVIDIA Vera Rubin and DSX Platform Advancements Showcase Energy Efficiencies of Optimizing Tokens Per Watt for AI Factories

NVIDIA presented its latest advances in energy-efficient AI infrastructure at the AI Infra Summit, focusing on maximizing AI tokens per megawatt of power rather than raw processing capability alone. Lambda improved performance per watt by 23 percent using NVIDIA's DSX MaxLPS software, and Emerald AI demonstrated how AI factories can adjust power consumption based on grid demand. NVIDIA DSX MaxLPS can deliver up to 1.4x more tokens per megawatt through factory-wide optimization and enable up to 40 percent more GPU capacity within the same power budget for Vera Rubin systems.

15 Sept blogs.nvidia.com

BlogNVIDIA

Perplexity Portable Computer Is Now Available on Windows, Powered by NVIDIA RTX

Perplexity Portable Computer is now available for Windows PCs with NVIDIA RTX graphics cards. It's a local AI agent that can plan and execute multi-step tasks directly on your computer without sending sensitive data to the cloud. The agent can analyze files, gather information from apps like Gmail, Slack, and GitHub, and handle work such as reviewing code changes, analyzing financial records, or finding where a startup loses new users. For more advanced tasks, it can itself request permission to use cloud-based AI. Setup is simple — the app uses pre-optimized AI models so you don't need to configure complex technology yourself.

14 Sept blogs.nvidia.com

BlogNVIDIA

Skild AI Taps NVIDIA Physical AI to Teach Robots New Tasks From a Single Video

Skild AI has launched S1, a robot model that can learn new tasks from just a video without requiring retraining. Instead of reprogramming robots for each new task, an operator can simply record a video of what needs to be done, and the model understands and performs it. S1 handles complex tasks lasting up to 10 minutes — such as plant potting, food preparation, and assembly work — and succeeds approximately 66 percent of the time on each step, compared to 9 percent for similar systems. Skild AI already has 60+ deployment partnerships and reached 100 million dollars in annual revenue 10 months after its first commercial deployment.

10 Sept blogs.nvidia.com

BlogNVIDIA

Physical AI Takes the Wheel: How the World’s Robotaxi Leaders Are Building With NVIDIA Technologies

NVIDIA presents a platform for robotaxi companies consisting of three computing components: one for training AI models, one for simulation and validation, and one for in-vehicle computations. The robotaxi market is estimated to grow to $400 billion by 2035 with over 6 million vehicles in operation, and all major commercial robotaxi programs today use NVIDIA's technology. The platform enables developers to train models on their own data, simulate millions of driving scenarios digitally, and run real-time computations safely in the vehicle—all to build and scale autonomous fleets.

10 Sept blogs.nvidia.com

BlogNVIDIA

d-Matrix Adopts NVIDIA NVLink Fusion for Rack-Scale XPU Deployment

Chipmaker d-Matrix will use NVLink Fusion, a standard from NVIDIA, to connect its Raptor XPU processors to NVIDIA's AI infrastructure. This enables d-Matrix to build large AI systems without solving all technical problems themselves—they can use NVIDIA's ready-made solutions for networking, server racks, cooling, and supply chains. Instead of building entirely custom servers, data centers can now use the same rack design for GPUs, CPUs, and XPUs together.

10 Sept blogs.nvidia.com

BlogNVIDIA

NVIDIA Brings Real-Time AI to Broadcast, Sports and Global Streaming at IBC

NVIDIA presented new AI tools for the media industry at the IBC conference in Amsterdam (September 11-14). The main initiative is an expansion of NVIDIA AI for Media, which includes the Synthetic Video Detector (SVD) for detecting deepfakes — the tool achieves 99.3 percent accuracy for text-to-video and 97.7 percent for image-to-video. Companies such as Dalet, TwelveLabs, and Wowza are integrating SVD into their platforms for news verification and content control. Additionally, 3D Body Pose was presented to track movement from a standard camera, which can be used in sports and animation production, along with tools for smoothing video motion, improving image quality, and converting standard video to HDR format.

9 Sept blogs.nvidia.com

BlogNVIDIA

Sparks Fly: NVIDIA Accelerates Local AI at IFA 2026

NVIDIA is presenting new tools at IFA 2026 that make it easier to run AI agents locally on your own computer instead of in the cloud. New RTX Spark computers from Lenovo and Acer will arrive in October, and popular agent apps like Hermes, OpenClaw, and Perplexity will get simplified local installation. NVIDIA has also released faster processing — up to 1.9 times faster — and launched several new AI models that can run locally, including Meta's Muse Glimmer and DeepSeek's v4 Flash. A new tool called NVIDIA PAIR intelligently distributes AI processing across multiple computers in your network.

3 Sept blogs.nvidia.com

BlogNVIDIA

NVIDIA to Acquire Hugging Face

NVIDIA is acquiring Hugging Face for approximately $12.9 billion. Hugging Face is a platform where over 18 million developers, researchers, and creators share more than 3 million AI models, 500,000 datasets, and 1 million applications—and over 200,000 companies use it. NVIDIA promises that Hugging Face will remain an open platform where developers can choose which models, frameworks, cloud providers, and hardware they want to use, and that NVIDIA's own hardware will not be required. NVIDIA states that open AI models give startups, universities, and small companies the opportunity to build on advanced AI capabilities without training models from scratch.

3 Sept blogs.nvidia.com

BlogNVIDIA

NVIDIA and CrowdStrike Strengthen Agentic Cybersecurity Frontier

NVIDIA and CrowdStrike introduced SafeMind, an AI-based defense system against cyberattacks that uses itself to improve continuously. The system is built on NVIDIA's Nemotron models trained on CrowdStrike's 15 years of security data, creating a loop where attack patterns and defenses challenge and enhance each other continuously. According to CrowdStrike, AI-enabled attacks increased 89 percent last year and hackers now take only 27 seconds to penetrate networks—faster than humans can respond. SafeMind is delivered built into CrowdStrike's Falcon platform and can also be used modularly, allowing organizations to choose which models and tools suit their environment.

1 Sept blogs.nvidia.com

BlogNVIDIA

NVIDIA NVLink Fusion Expands With NVHBM Custom High-Bandwidth Memory

NVIDIA has expanded its NVLink Fusion platform with NVHBM, a new high-speed memory technology. Instead of placing the memory controller on the AI chip itself (XPU), NVHBM places it in the memory stack, delivering 30 percent higher memory speed, 15 percent lower power consumption, and freeing up 25 percent more space for computation on the chip. Amazon will be the first to use the technology with its future Trainium4 chips.

26 Aug blogs.nvidia.com

BlogNVIDIA

Leading Publishers Bring Blockbuster PC Games and Technology to NVIDIA RTX Spark

NVIDIA is launching RTX Spark, a new platform for Windows PCs arriving in fall that combines AI agents, content creation, and high-performance gaming. Major game publishers including EA, Ubisoft, and Embark Studios are bringing popular titles to the platform, such as EA SPORTS F1 25, Apex Legends, and Anno 117: Pax Romana. NVIDIA also announces updates to its graphics technologies, including DLSS 4.5 Ray Reconstruction, which uses AI to improve image quality, and support for EA's anti-cheat technology to enable fair play.

25 Aug blogs.nvidia.com

BlogNVIDIA

How XPUs Meet a World-Class AI Factory

NVIDIA introduces NVLink Fusion, a solution that allows companies to build custom AI processors (XPU) without developing all infrastructure from scratch. Instead of designing networks, server racks, and cooling equipment independently, companies can connect their custom processors to NVIDIA's existing systems. NVLink Fusion delivers faster communication between processors — approximately 3 times lower latency than standard networking alternatives — and supports up to 72 processors in a network. It also enables companies to prepare their data centers before the final processor is ready, reducing risk and time to market.

24 Aug blogs.nvidia.com

BlogNVIDIA

With Groq 3 LPX in Full Production, NVIDIA Extends Vera Rubin Inference for Agents

NVIDIA has launched Groq 3 LPX, a system now in full production designed to run agent-based AI systems that can make independent decisions and use tools. The system can generate 3,400 output tokens per second—approximately four times faster than the nearest competitor's solution. NVIDIA combines three components: Vera Rubin processors for reading and processing large amounts of text, Spectrum-X networks for fast data communication, and Groq 3 LPX for generating responses. Companies including SpaceXAI, CoreWeave, and Nebius are already adopting the platform.

24 Aug blogs.nvidia.com

BlogNVIDIA

Up to 30x More Work Per Watt: NVIDIA Vera Rubin NVL72 Sets a New Efficiency Standard for AI Agents

NVIDIA presents its new Vera Rubin NVL72 processor, which handles AI agents 30 times more efficiently than previous models. Agents consume much more data than simple chat — they search through databases, compare alternatives, and perform reasoning step by step, causing the volume of processed data to grow enormously. Vera Rubin also reduces cost per million processed tokens by 35 times. This matters for companies running AI factories, as it enables running more agents on the same power budget and earning more money per processed request.

24 Aug blogs.nvidia.com

← All news