NVIDIA Announces DGX Spark and DGX Station Personal AI Computers
NVIDIA has unveiled two new personal AI supercomputers: DGX Spark and DGX Station, powered by the NVIDIA Grace Blackwell platform. These desktop systems bring data-center-level AI capabilities to developers, researchers, and data scientists.
DGX Spark, featuring the GB10 Grace Blackwell Superchip, delivers up to 1,000 trillion operations per second of AI compute. The system includes fifth-generation Tensor Cores and FP4 support, optimized for AI model fine-tuning and inference.
DGX Station, built with the GB300 Grace Blackwell Ultra Desktop Superchip, offers 784GB of coherent memory space and includes the ConnectX-8 SuperNIC supporting networking up to 800Gb/s. Leading computer manufacturers including ASUS, Dell Technologies, HP, and Lenovo will develop these systems.
Reservations for DGX Spark are now open, while DGX Station will be available from manufacturing partners later in 2025.
NVIDIA ha presentato due nuovi supercomputer personali per l'IA: DGX Spark e DGX Station, alimentati dalla piattaforma NVIDIA Grace Blackwell. Questi sistemi desktop portano capacità di intelligenza artificiale a livello di data center a sviluppatori, ricercatori e scienziati dei dati.
Il DGX Spark, dotato del Superchip GB10 Grace Blackwell, offre fino a 1.000 trilioni di operazioni al secondo per il calcolo dell'IA. Il sistema include i Tensor Core di quinta generazione e supporto FP4, ottimizzati per la messa a punto e l'inferenza dei modelli di IA.
Il DGX Station, costruito con il Superchip GB300 Grace Blackwell Ultra Desktop, offre 784GB di spazio di memoria coerente e include il ConnectX-8 SuperNIC che supporta reti fino a 800Gb/s. I principali produttori di computer, tra cui ASUS, Dell Technologies, HP e Lenovo, svilupperanno questi sistemi.
Le prenotazioni per il DGX Spark sono ora aperte, mentre il DGX Station sarà disponibile dai partner produttori entro la fine del 2025.
NVIDIA ha presentado dos nuevos supercomputadores personales de IA: DGX Spark y DGX Station, impulsados por la plataforma NVIDIA Grace Blackwell. Estos sistemas de escritorio llevan capacidades de IA a nivel de centro de datos a desarrolladores, investigadores y científicos de datos.
El DGX Spark, que cuenta con el Superchip GB10 Grace Blackwell, ofrece hasta 1.000 billones de operaciones por segundo en computación de IA. El sistema incluye núcleos Tensor de quinta generación y soporte FP4, optimizados para el ajuste fino e inferencia de modelos de IA.
El DGX Station, construido con el Superchip GB300 Grace Blackwell Ultra Desktop, ofrece 784GB de espacio de memoria coherente e incluye el ConnectX-8 SuperNIC que soporta redes de hasta 800Gb/s. Los principales fabricantes de computadoras, incluidos ASUS, Dell Technologies, HP y Lenovo, desarrollarán estos sistemas.
Las reservas para el DGX Spark están ahora abiertas, mientras que el DGX Station estará disponible de los socios de fabricación a finales de 2025.
NVIDIA는 개인 AI 슈퍼컴퓨터 두 대인 DGX Spark와 DGX Station을 발표했습니다. 이들은 NVIDIA Grace Blackwell 플랫폼으로 구동됩니다. 이 데스크탑 시스템은 개발자, 연구원 및 데이터 과학자에게 데이터 센터 수준의 AI 기능을 제공합니다.
GB10 Grace Blackwell 슈퍼칩을 탑재한 DGX Spark는 초당 최대 1,000조 개의 AI 연산을 제공합니다. 이 시스템은 AI 모델의 미세 조정 및 추론을 위해 최적화된 5세대 텐서 코어와 FP4 지원을 포함합니다.
GB300 Grace Blackwell Ultra Desktop 슈퍼칩으로 제작된 DGX Station은 784GB의 일관된 메모리 공간을 제공하며, 최대 800Gb/s의 네트워킹을 지원하는 ConnectX-8 SuperNIC이 포함되어 있습니다. ASUS, Dell Technologies, HP, Lenovo를 포함한 주요 컴퓨터 제조업체들이 이 시스템을 개발할 것입니다.
DGX Spark의 예약이 이제 열렸으며, DGX Station은 2025년 후반에 제조 파트너를 통해 제공될 예정입니다.
NVIDIA a dévoilé deux nouveaux superordinateurs personnels d'IA : DGX Spark et DGX Station, alimentés par la plateforme NVIDIA Grace Blackwell. Ces systèmes de bureau apportent des capacités d'IA de niveau centre de données aux développeurs, chercheurs et scientifiques des données.
Le DGX Spark, doté de la Superpuce GB10 Grace Blackwell, offre jusqu'à 1 000 billions d'opérations par seconde pour le calcul d'IA. Le système comprend des cœurs Tensor de cinquième génération et un support FP4, optimisés pour le réglage fin et l'inférence des modèles d'IA.
La DGX Station, construite avec la Superpuce GB300 Grace Blackwell Ultra Desktop, offre 784 Go d'espace mémoire cohérent et inclut le ConnectX-8 SuperNIC qui prend en charge des réseaux allant jusqu'à 800 Gb/s. Les principaux fabricants d'ordinateurs, y compris ASUS, Dell Technologies, HP et Lenovo, développeront ces systèmes.
Les réservations pour le DGX Spark sont désormais ouvertes, tandis que le DGX Station sera disponible auprès des partenaires de fabrication d'ici fin 2025.
NVIDIA hat zwei neue persönliche KI-Supercomputer vorgestellt: DGX Spark und DGX Station, die von der NVIDIA Grace Blackwell Plattform betrieben werden. Diese Desktop-Systeme bringen KI-Funktionalitäten auf Rechenzentrumsniveau zu Entwicklern, Forschern und Datenwissenschaftlern.
Der DGX Spark, ausgestattet mit dem GB10 Grace Blackwell Superchip, bietet bis zu 1.000 Billionen Operationen pro Sekunde für KI-Berechnungen. Das System umfasst Tensor Cores der fünften Generation und FP4-Unterstützung, die für das Feintuning und die Inferenz von KI-Modellen optimiert sind.
Die DGX Station, die mit dem GB300 Grace Blackwell Ultra Desktop Superchip gebaut wurde, bietet 784 GB zusammenhängenden Speicherplatz und enthält das ConnectX-8 SuperNIC, das Netzwerke mit bis zu 800 Gb/s unterstützt. Führende Computerhersteller wie ASUS, Dell Technologies, HP und Lenovo werden diese Systeme entwickeln.
Die Reservierungen für den DGX Spark sind jetzt geöffnet, während die DGX Station später im Jahr 2025 von den Fertigungspartnern verfügbar sein wird.
- Launch of two new high-performance AI supercomputers for desktop use
- DGX Spark delivers 1,000 trillion operations per second of AI compute
- DGX Station features substantial 784GB coherent memory space
- Partnership with major manufacturers (ASUS, Dell, HP, Lenovo) for distribution
- High-speed networking capability up to 800Gb/s with ConnectX-8 SuperNIC
- DGX Station's delayed availability until later in 2025
Insights
NVIDIA's latest product launch represents a significant strategic expansion of its AI computing portfolio, bringing data center-grade technology to desktop form factors. The new DGX Spark and DGX Station personal AI supercomputers powered by the Grace Blackwell platform target a growing market of AI developers, researchers, and data scientists who need local computing power for model development.
This move effectively extends NVIDIA's addressable market by creating a new product category between consumer GPUs and data center solutions. By partnering with major OEMs including ASUS, Dell, HP, and Lenovo for distribution, NVIDIA ensures broad market reach while maintaining its premium positioning.
The GB10 and GB300 Grace Blackwell chips powering these systems showcase NVIDIA's technological leadership, particularly the NVLink-C2C interconnect that delivers 5x the bandwidth of fifth-generation PCIe. This creates a significant performance moat against potential competitors.
Most strategically valuable is how these products strengthen NVIDIA's full-stack AI ecosystem. The seamless workflow from desktop development to DGX Cloud deployment creates customer lock-in and potentially drives increased cloud service adoption, supporting NVIDIA's recurring revenue ambitions.
While pricing details aren't disclosed, these will likely be premium products with healthy margins, following NVIDIA's established business model of commanding price premiums for performance leadership.
NVIDIA's new DGX personal AI computers represent a critical technological bridge between consumer and enterprise AI computing. The technical specifications reveal NVIDIA's significant architectural advantages, with the GB10 Superchip delivering up to 1,000 trillion operations per second of AI compute optimized for the latest reasoning models.
The coherent memory model between CPU and GPU via NVLink-C2C interconnect solves one of the most significant bottlenecks in AI development workloads - memory bandwidth. This advancement allows developers to work with larger models locally, accelerating development cycles and reducing dependency on cloud resources for iteration.
The DGX Station's massive 784GB of coherent memory space is particularly notable for enabling local training of substantial models that previously required data center resources. The inclusion of the ConnectX-8 SuperNIC with up to 800Gb/s networking creates additional flexibility for clustering multiple units.
By bundling these systems with NVIDIA AI Enterprise software and NIM microservices, NVIDIA continues its strategy of offering optimized full-stack solutions rather than mere hardware components. This approach significantly raises switching costs for developers who build workflows around NVIDIA's software ecosystem.
These products essentially democratize access to AI supercomputing capabilities that were previously restricted to enterprises with substantial infrastructure budgets, potentially accelerating AI application development across industries.
Powered by NVIDIA Grace Blackwell, Desktop Supercomputers Place Accelerated AI in the Hands of Developers, Researchers and Data Scientists; Systems Coming From Leading Computer Makers Including ASUS, Dell Technologies, HP and Lenovo
SAN JOSE, Calif., March 18, 2025 (GLOBE NEWSWIRE) -- GTC -- NVIDIA today unveiled NVIDIA DGX™ personal AI supercomputers powered by the NVIDIA Grace Blackwell platform.
DGX Spark — formerly Project DIGITS — and DGX Station™, a new high-performance NVIDIA Grace Blackwell desktop supercomputer powered by the NVIDIA Blackwell Ultra platform, enable AI developers, researchers, data scientists and students to prototype, fine-tune and inference large models on desktops. Users can run these models locally or deploy them on NVIDIA DGX Cloud or any other accelerated cloud or data center infrastructure.
DGX Spark and DGX Station bring the power of the Grace Blackwell architecture, previously only available in the data center, to the desktop. Global system builders to develop DGX Spark and DGX Station include ASUS, Dell, HP Inc. and Lenovo.
“AI has transformed every layer of the computing stack. It stands to reason a new class of computers would emerge — designed for AI-native developers and to run AI-native applications,” said Jensen Huang, founder and CEO of NVIDIA. “With these new DGX personal AI computers, AI can span from cloud services to desktop and edge applications.”
Igniting Innovation With DGX Spark
DGX Spark is the world’s smallest AI supercomputer, empowering millions of researchers, data scientists, robotics developers and students to push the boundaries of generative and physical AI with massive performance and capabilities.
At the heart of DGX Spark is the NVIDIA GB10 Grace Blackwell Superchip, optimized for a desktop form factor. GB10 features a powerful NVIDIA Blackwell GPU with fifth-generation Tensor Cores and FP4 support, delivering up to 1,000 trillion operations per second of AI compute for fine-tuning and inference with the latest AI reasoning models, including the NVIDIA Cosmos Reason world foundation model and NVIDIA GR00T N1 robot foundation model.
The GB10 Superchip uses NVIDIA NVLink™-C2C interconnect technology to deliver a CPU+GPU-coherent memory model with 5x the bandwidth of fifth-generation PCIe. This lets the superchip access data between a GPU and CPU to optimize performance for memory-intensive AI developer workloads.
NVIDIA’s full-stack AI platform enables DGX Spark users to seamlessly move their models from their desktops to DGX Cloud or any accelerated cloud or data center infrastructure — with virtually no code changes — making it easier than ever to prototype, fine-tune and iterate on their workflows.
Full Speed Ahead With DGX Station
NVIDIA DGX Station brings data-center-level performance to desktops for AI development. The first desktop system to be built with the NVIDIA GB300 Grace Blackwell Ultra Desktop Superchip, DGX Station features a massive 784GB of coherent memory space to accelerate large-scale training and inferencing workloads. The GB300 Desktop Superchip features an NVIDIA Blackwell Ultra GPU with latest-generation Tensor Cores and FP4 precision — connected to a high-performance NVIDIA Grace™ CPU via NVLink-C2C — delivering best-in-class system communication and performance.
DGX Station also features the NVIDIA ConnectX®-8 SuperNIC, optimized to supercharge hyperscale AI computing workloads. With support for networking at up to 800Gb/s, the ConnectX-8 SuperNIC delivers extremely fast, efficient network connectivity, enabling high-speed connectivity of multiple DGX Stations for even larger workloads, and network-accelerated data transfers for AI workloads.
Combining these state-of-the-art DGX Station capabilities with the NVIDIA CUDA-X™ AI platform, teams can achieve exceptional desktop AI development performance.
In addition, users gain access to NVIDIA NIM™ microservices with the NVIDIA AI Enterprise software platform, which offers highly optimized, easy-to-deploy inference microservices backed by enterprise support.
Availability
Reservations for DGX Spark systems open today at nvidia.com.
DGX Station is expected to be available from manufacturing partners like ASUS, BOXX, Dell, HP, Lambda and Supermicro later this year.
Learn more by watching the NVIDIA GTC keynote and register for sessions from NVIDIA and industry leaders at the show, which runs through March 21.
About NVIDIA
NVIDIA (NASDAQ: NVDA) is the world leader in accelerated computing.
For further information, contact:
Pearlina Boc
NVIDIA Corporation
+1-562-275-5781
pboc@nvidia.com
Certain statements in this press release including, but not limited to, statements as to: the benefits, impact, availability, and performance of NVIDIA’s products, services, and technologies; third parties adopting or offering NVIDIA’s products and technologies; by putting the NVIDIA Grace Blackwell Superchip on every desk, and at every AI developer’s fingertips, NVIDIA empowering millions of people to shape the future of AI; and with new DGX AI supercomputers, software providers, government agencies, startups and researchers being able to prototype, fine-tune and run large AI models — transforming the way they work and create are forward-looking statements that are subject to risks and uncertainties that could cause results to be materially different than expectations. Important factors that could cause actual results to differ materially include: global economic conditions; our reliance on third parties to manufacture, assemble, package and test our products; the impact of technological development and competition; development of new products and technologies or enhancements to our existing product and technologies; market acceptance of our products or our partners' products; design, manufacturing or software defects; changes in consumer preferences or demands; changes in industry standards and interfaces; unexpected loss of performance of our products or technologies when integrated into systems; as well as other factors detailed from time to time in the most recent reports NVIDIA files with the Securities and Exchange Commission, or SEC, including, but not limited to, its annual report on Form 10-K and quarterly reports on Form 10-Q. Copies of reports filed with the SEC are posted on the company's website and are available from NVIDIA without charge. These forward-looking statements are not guarantees of future performance and speak only as of the date hereof, and, except as required by law, NVIDIA disclaims any obligation to update these forward-looking statements to reflect future events or circumstances.
© 2025 NVIDIA Corporation. All rights reserved. NVIDIA, the NVIDIA logo, Connect-X, CUDA-X, DGX, DGX Station, NVIDIA Grace, NVIDIA NIM and NVLink are trademarks and/or registered trademarks of NVIDIA Corporation in the U.S. and/or other countries. Other company and product names may be trademarks of the respective companies with which they are associated. Features, pricing, availability, and specifications are subject to change without notice.
A photo accompanying this announcement is available at https://www.globenewswire.com/NewsRoom/AttachmentNg/a1933d3b-32bd-450a-88f9-cd3df95857e1
FAQ
What are the key features of NVIDIA's new DGX Spark AI supercomputer (NVDA)?
How much memory does the NVIDIA DGX Station (NVDA) offer for AI workloads?
Which manufacturers will produce NVIDIA's new DGX personal AI computers (NVDA)?
What is the network performance capability of NVIDIA's DGX Station (NVDA)?
When will NVIDIA's new DGX personal AI computers be available (NVDA)?
