Description
- Cortex-X925 10-Core CPU
- Cortex-A725 10-Core CPU
- 128GB of LPDDR5x RAM
- 4TB NVMe M.2 PCIe 4.0 SSD
NVIDIA DGX Spark Desktop Computer
Feature
- Built on NVIDIA GB10 Grace Blackwell Superchip
- NVIDIA Blackwell GPU with fifth-generation Tensor Core technology
- NVIDIA Grace CPU with 20-core high-performance Arm architecture
- Up to 1 petaFLOP1 of AI performance using FP4
- 128 GB of coherent, unified system memory
- Support for up to 200 billion parameter2 models
- NVIDIA ConnectX networking to link two systems to work with models up to 405 billion parameters
- 4 TB of NVMe storage
- Compact desktop form factor
Featuring a combined Cortex-X925 10-Core and Cortex-A725 10-Core processor, 128GB of unified LPDDR5x memory, and a 4TB M.2 NVMe PCIe 4.0 SSD, the DGX Spark is ready to handle a wide range of workloads with the preinstalled NVIDIA AI software stack – including prototyping, fine-tuning and inference, and data science applications. Four USB 3.2 Gen 2×2 ports are available, one of which serves as a power input for the included 240W adapter while the other three work with up to 20 Gb/s compatible peripherals. There is also an HDMI 2.1a port for external displays. Wi-Fi 7 (802.11be) support and a 10 Gb/s RJ45 Ethernet port provide high-speed connections to compatible networks. Two 200 Gb/s QSFP112 ports on the NVIDIA ConnectX-7 SmartNIC adapter allow you to scale performance by linking up to two Ascent GX10 systems together to handle large models with 405 billion parameters. The NVIDIA DGX OS operating system is preinstalled.Your Personal AI Supercomputer
Breakthrough AI Architecture
The NVIDIA GB10 Grace Blackwell Superchip and fifth-generation Tensor Cores drive cutting-edge performance and reliability in a compact design. The AI TOP ATOM also features NVIDIA NVLink-C2C technology with a 20-Core ARM processor that supports up to 200 billion parameter models.Accelerate All AI Workloads
Prototyping – Develop, test, and validate AI models and applications
Fine-Tuning – Fine-tune AI models with up to 70 billion parameters.
Inference – Test, validate, and inference with AI models up to 200 billion parameters.
Data Science – Accomplish high-performance data science tasks at your desk.
Edge Application – Develop edge applications with NVIDIA AI frameworks, including Isaac, Metropolis, and many others.
NVIDIA DGX Spark Software
NVIDIA DGX Spark software stack accelerates AI and deep learning workloads while streamlining prototyping and development. It provides a scalable foundation for AI initiatives, enabling developers to harness NVIDIA’s powerful infrastructure to drive innovation and achieve results.
Technical Specifications
| Brand | NVIDIA |
| Architecture | NVIDIA Grace Blackwell |
| GPU | Blackwell Architecture |
| CPU | 20 core Arm, 10 Cortex-X925 + 10 Cortex-A725 Arm |
| CUDA Cores | Blackwell Generation |
| Tensor Cores | 5th Generation |
| RT Cores | 4th Generation |
| Tensor Performance1 | 1 PFLOP |
| System Memory | 128 GB LPDDR5x, coherent unified system memory |
| Memory Interface | 256-bit |
| Memory Bandwidth | 273 GB/s |
| Storage | 4 TB NVME.M2 with self-encryption |
| USB | 4x USB TypeC |
| Ethernet | 1x RJ-45 connector |
| NIC | ConnectX-7 NIC @ 200 Gbps |
| Wi-Fi | WiFi 7 |
| Bluetooth | BT 5.4 |
| Audio-output | HDMI multichannel audio output |
| Power Supply | 240 W |
| GB10 TDP2 | 140 W |
| Display Connectors | 1x HDMI 2.1a |
| NVENC | NVDEC | 1x | 1x |
| OS | NVIDIA DGX™ OS |
| System Dimensions | 150 mm L x 150 mm W x 50.5 mm H |
| System Weight | 1.2 kg |
Note: Product features, specifications, and physical appearance are subject to change without notice due to manufacturer updates. Please verify compatibility with your system before purchase.
Customer Reviews
There are no reviews yet.