NVIDIA DGX Spark Arrives for World’s AI Developers
NVIDIA (NVDA) began shipping DGX Spark, described as the world’s smallest AI supercomputer, with orders available starting Oct. 15, 2025.
DGX Spark delivers up to 1 petaflop of AI performance and 128GB unified CPU–GPU memory, built on the NVIDIA Grace Blackwell architecture with GB10 Superchip, ConnectX-7 200 Gb/s networking and NVLink‑C2C. It supports local inference of models up to 200 billion parameters and fine‑tuning of models up to 70 billion parameters. Systems are offered through OEM partners and select retail channels worldwide, with the NVIDIA AI software stack preinstalled for immediate developer use.
NVIDIA (NVDA) ha iniziato a spedire DGX Spark, descritto come il supercomputer AI più piccolo al mondo, con ordini disponibili dal 15 ottobre 2025.
DGX Spark offre fino a 1 petaflop di prestazioni AI e 128GB di memoria unificata CPU–GPU, costruita sull'architettura NVIDIA Grace Blackwell con GB10 Superchip, ConnectX-7 200 Gb/s networking e NVLink‑C2C. Supporta l'inferenza locale di modelli fino a 200 miliardi di parametri e il fine‑tuning di modelli fino a 70 miliardi di parametri. I sistemi sono offerti tramite partner OEM e canali retail selezionati in tutto il mondo, con lo stack software NVIDIA AI preinstallato per un immediato utilizzo da parte degli sviluppatori.
NVIDIA (NVDA) comenzó a distribuir DGX Spark, descrito como la supercomputadora de IA más pequeña del mundo, con pedidos disponibles a partir del 15 de octubre de 2025.
DGX Spark ofrece hasta 1 petaflop de rendimiento de IA y 128GB de memoria unificada CPU–GPU, basada en la arquitectura NVIDIA Grace Blackwell con GB10 Superchip, conectividad de red ConnectX-7 200 Gb/s y NVLink‑C2C. Admite la inferencia local de modelos de hasta 200 mil millones de parámetros y el ajuste fino de modelos de hasta 70 mil millones de parámetros. Los sistemas se ofrecen a través de socios OEM y canales minoristas selectos en todo el mundo, con la pila de software NVIDIA AI preinstalada para uso inmediato por parte de los desarrolladores.
NVIDIA (NVDA)는 세계에서 가장 작은 AI 슈퍼컴퓨터로 묘사된 DGX Spark의 배송을 시작했으며 주문은 2025년 10월 15일부터 가능하다.
DGX Spark은 최대 1 페타플롭의 AI 성능과 128GB 통합 CPU–GPU 메모리를 제공하며 NVIDIA Grace Blackwell 아키텍처에 GB10 슈퍼칩, ConnectX-7 200 Gb/s 네트워킹 및 NVLink-C2C로 구성된다. 로컬 추론은 최대 2000억 개의 파라미터 모델까지, 파인튜닝은 최대 700억 개의 파라미터까지 지원한다. 시스템은 OEM 파트너 및 선정된 소매 채널을 통해 전 세계에 제공되며, 개발자 즉시 사용 가능하도록 NVIDIA AI 소프트웨어 스택이 사전 설치되어 있다.
NVIDIA (NVDA) a commencé à expédier DGX Spark, décrit comme le plus petit superordinateur IA au monde, avec des commandes disponibles à partir du 15 octobre 2025.
DGX Spark offre jusqu'à 1 petaflop de performance IA et 128 Go mémoire unifiée CPU–GPU, construit sur l'architecture NVIDIA Grace Blackwell avec le GB10 Superchip, le réseau ConnectX-7 200 Gb/s et NVLink-C2C. Il prend en charge l'inférence locale de modèles jusqu'à 200 milliards de paramètres et l'ajustement fin de modèles jusqu'à 70 milliards de paramètres. Les systèmes sont proposés via des partenaires OEM et des canaux de détail choisis dans le monde entier, avec la pile logicielle NVIDIA AI préinstallée pour une utilisation immédiate par les développeurs.
NVIDIA (NVDA) hat den Versand von DGX Spark begonnen, der als weltweit kleinster AI-Supercomputer beschrieben wird, Bestellungen sind ab dem 15. Oktober 2025 möglich.
DGX Spark bietet bis zu 1 Petaflop KI-Leistung und 128 GB einheitlichen CPU–GPU-Speicher, basierend auf der NVIDIA Grace Blackwell-Architektur mit GB10 Superchip, ConnectX-7 200 Gb/s Networking und NVLink-C2C. Es unterstützt lokale Inferenz von Modellen bis zu 200 Milliarden Parametern und Feineinstellung von Modellen bis zu 70 Milliarden Parametern. Systeme werden weltweit über OEM-Partner und ausgewählte Einzelhandelskanäle angeboten, mit dem NVIDIA AI-Softwarestack vorinstalliert für sofortige Entwicklernutzung.
NVIDIA (NVDA) بدأت شحن DGX Spark، الذي يوصف بأنه أصغر كمبيوتر فائق للذكاء الاصطناعي في العالم، مع إمكانية الطلبات اعتبارًا من 15 أكتوبر 2025.
يقدم DGX Spark أداء ذكاء اصطناعي يصل إلى 1 فتر فيلوب؟ نصح مسجد: الترجمة الصحيحة هي 1 Petaflop. 128GB ذاكرة موحدة CPU–GPU، مبنية على بنية NVIDIA Grace Blackwell مع GB10 Superchip، شبكات ConnectX-7 بسرعة 200 Gb/s و NVLink-C2C. يدعم الاستدلال المحلي للنماذج حتى 200 مليار بارامتر وضبط النماذج حتى 70 مليار بارامتر. الأنظمة تُعرض من خلال شركاء OEM وقنوات التجزئة المختارة حول العالم، مع مجموعة NVIDIA AI البرمجية المثبتة مسبقاً لاستخدام المطورين فوراً.
NVIDIA (NVDA) 已开始出货 DGX Spark，被描述为全球最小的人工智能超级计算机，订单自 2025 年 10 月 15 日起接受。
DGX Spark 提供高达 1 petaflop 的 AI 性能和 128GB 统一的 CPU–GPU 内存，基于 NVIDIA Grace Blackwell 架构，搭载 GB10 超级芯片、200 Gb/s 的 ConnectX-7 网络以及 NVLink-C2C。它支持对最多 2000亿 参数的模型进行本地推理，以及对最多 700亿 参数的模型进行微调。系统通过 OEM 合作伙伴和全球精选零售渠道提供，预装 NVIDIA AI 软件栈，供开发者立即使用。
- 1 petaflop peak AI performance
- 128GB unified CPU–GPU coherent memory
- Supports inference of models up to 200B parameters
- Supports fine‑tuning of models up to 70B parameters
- Available from major partners and retail channels starting Oct 15, 2025
- Preinstalled NVIDIA AI software stack for out‑of‑the‑box development
- None.
Insights
DGX Spark ships worldwide, embedding NVIDIA's AI stack into a desktop supercomputer and broadening developer access.
As a compact system, DGX Spark packages NVIDIA GPUs, CPUs, networking, CUDA libraries, and NVIDIA AI software into a desktop form factor. It delivers up to 1 petaflop of AI performance and 128GB of unified memory, enabling local inference on models up to 200 billion parameters and fine‑tuning up to 70 billion parameters. Major OEM partners including Acer, ASUS, Dell Technologies, GIGABYTE, HP, Lenovo, and MSI will ship partner systems, with orders available starting
The main dependency is adoption of the new desktop workflow by developers and ISVs that must validate and optimize tools for this hardware. Early testing by platforms such as Hugging Face, Google, Microsoft, and others reduces integration risk, but broad software optimization remains required for wide uptake. Watch partner availability, channel inventory, and developer support bundles over the next 3–12 months to gauge real market traction and ecosystem momentum.
News Summary:
- NVIDIA founder and CEO Jensen Huang delivers DGX Spark to Elon Musk at SpaceX.
- This week, NVIDIA and its partners are shipping DGX Spark, the world’s smallest AI supercomputer, delivering NVIDIA’s AI stack in a compact desktop form factor.
- Acer, ASUS, Dell Technologies, GIGABYTE, HPI, Lenovo and MSI debut DGX Spark systems, expanding access to powerful AI computing.
- Built on the NVIDIA Grace Blackwell architecture, DGX Spark integrates NVIDIA GPUs, CPUs, networking, CUDA libraries and NVIDIA AI software, accelerating agentic and physical AI development.
SANTA CLARA, Calif., Oct. 13, 2025 (GLOBE NEWSWIRE) -- NVIDIA today announced it will start shipping NVIDIA DGX Spark™, the world’s smallest AI supercomputer.
AI workloads are quickly outgrowing the memory and software capabilities of the PCs, workstations and laptops millions of developers rely on today — forcing teams to shift work to the cloud or local data centers.
As a new class of computer, DGX Spark delivers a petaflop of AI performance and 128GB of unified memory in a compact desktop form factor, giving developers the power to run inference on AI models with up to 200 billion parameters and fine-tune models of up to 70 billion parameters locally. In addition, DGX Spark lets developers create AI agents and run advanced software stacks locally.
“In 2016, we built DGX-1 to give AI researchers their own supercomputer. I hand-delivered the first system to Elon at a small startup called OpenAI — and from it came ChatGPT, kickstarting the AI revolution,” said Jensen Huang, founder and CEO of NVIDIA. “DGX-1 launched the era of AI supercomputers and unlocked the scaling laws that drive modern AI. With DGX Spark, we return to that mission — placing an AI computer in the hands of every developer to ignite the next wave of breakthroughs.”
DGX Spark brings together the full NVIDIA AI platform — including GPUs, CPUs, networking, CUDA® libraries and the NVIDIA AI software stack — into a system small enough for a lab or an office, yet powerful enough to accelerate agentic and physical AI development. By combining breakthrough performance with the reach of the NVIDIA ecosystem, DGX Spark transforms the desktop into an AI development platform.
DGX Spark systems deliver up to 1 petaflop of AI performance, accelerated by a NVIDIA GB10 Grace Blackwell Superchip, NVIDIA ConnectX®-7 200 Gb/s networking and NVIDIA NVLink™-C2C technology, providing 5x the bandwidth of fifth-generation PCIe with 128GB of CPU-GPU coherent memory.
The NVIDIA AI software stack is preinstalled to enable developers to start working on AI projects out of the box. With DGX Spark, developers can access NVIDIA AI ecosystem tools including models, libraries, and NVIDIA NIM™ microservices, enabling local workflows such as customizing Black Forest Labs’ FLUX.1 models to refine image generation, creating a vision search and summarization agent using the NVIDIA Cosmos™ Reason vision language model, or building an AI chatbot using Qwen3 that is optimized for DGX Spark.
To celebrate DGX Spark shipping worldwide, Huang hand-delivered one of the first units of DGX Spark to Elon Musk, chief engineer at SpaceX, today in Starbase, Texas. The exchange was a connection to the supercomputer’s origins, as Musk was among the team that received the first NVIDIA DGX™-1 supercomputer from Huang in 2016.
Other early recipients of DGX Spark, including Anaconda, Cadence, ComfyUI, Docker, Google, Hugging Face, JetBrains, LM Studio, Meta, Microsoft, Ollama and Roboflow, are testing, validating and optimizing their tools, software and models for DGX Spark.
Research organizations around the world, including the NYU Global Frontier Lab, previewed DGX Spark to boost their AI development.
“DGX Spark allows us to access peta-scale computing on our desktop,” said Kyunghyun Cho, professor of computer and data science at the NYU Global Frontier Lab. “This new way to conduct AI research and development enables us to rapidly prototype and experiment with advanced AI algorithms and models — even for privacy- and security-sensitive applications, such as healthcare.”
NVIDIA DGX Spark Now Shipping
Starting Wednesday, Oct. 15, DGX Spark can be ordered on NVIDIA.com. Partner systems will be available from Acer, ASUS, Dell Technologies, GIGABYTE, HP, Lenovo, MSI as well as Micro Center stores in the U.S., and from NVIDIA channel partners worldwide.
About NVIDIA
NVIDIA (NASDAQ: NVDA) is the world leader in AI and accelerated computing.
For further information, contact:
Alex Shapiro
Public Relations
NVIDIA Corporation
+1-415-608-5044
ashapiro@nvidia.com
Certain statements in this press release including, but not limited to, statements as to: with DGX Spark, NVIDIA placing an AI computer in the hands of every developer to ignite the next wave of breakthroughs; the benefits, impact, performance, and availability of NVIDIA’s products, services, and technologies; expectations with respect to NVIDIA’s third party arrangements, including with its collaborators and partners; expectations with respect to technology developments; and other statements that are not historical facts are forward-looking statements within the meaning of Section 27A of the Securities Act of 1933, as amended, and Section 21E of the Securities Exchange Act of 1934, as amended, which are subject to the “safe harbor” created by those sections based on management’s beliefs and assumptions and on information currently available to management and 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 and political conditions; NVIDIA’s reliance on third parties to manufacture, assemble, package and test NVIDIA’s products; the impact of technological development and competition; development of new products and technologies or enhancements to NVIDIA’s existing product and technologies; market acceptance of NVIDIA’s products or NVIDIA’s partners’ products; design, manufacturing or software defects; changes in consumer preferences or demands; changes in industry standards and interfaces; unexpected loss of performance of NVIDIA’s products or technologies when integrated into systems; and changes in applicable laws and regulations, 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, ConnectX, CUDA, DGX, NVIDIA Cosmos, NVIDIA NIM and NVLink are trademarks and/or registered trademarks of NVIDIA Corporation in the U.S. and other countries. Other NVIDIA trademarks and service marks not listed herein are also the property of NVIDIA Corporation. All rights reserved. All other company and product names are trademarks or registered trademarks of their respective owners. Features, pricing, availability and specifications are subject to change without notice.
Photos accompanying this announcement are available at
https://www.globenewswire.com/NewsRoom/AttachmentNg/ee138c12-fd60-431a-9b43-c6fb1a34287c
https://www.globenewswire.com/NewsRoom/AttachmentNg/a6e9e492-6a02-47d9-9042-856c567fd7c9
FAQ
When can developers order NVIDIA DGX Spark (NVDA) and where is it sold?
What performance and memory does DGX Spark (NVDA) provide for local AI work?
What model sizes can DGX Spark (NVDA) run for inference and fine‑tuning?
Which NVIDIA technologies are integrated into DGX Spark (NVDA)?
Will DGX Spark require cloud or data‑center workflows for agentic AI development?
