ASUS Ascent GX10: AI supercomputer based on the NVIDIA DGX Spark platform

The ASUS Ascent GX10 is a compact AI supercomputer. A generous 128GB of RAM enables work with large LLMs with up to 200B parameters.

The core of the ASUS Ascent GX10 is the NVIDIA® GB10 Grace Blackwell superchip, with a performance of 1 petaflop for AI calculations (in FP4 precision).

The NVIDIA DGX Spark platform is especially well-suited for AI developers and AI research. Thanks to its compact form factor, the ASUS Ascent GX10 fits perfectly on any desktop.

A key component of the platform is the NVIDIA AI software stack, which enables a quick start with the best-supported platform in the LLM industry. Thanks to software compatibility with NVIDIA's larger platforms (e.g., H200 / B200), the developed LLM solutions and applications (e.g., agentic AI) can be scaled directly and easily.

Thanks to the built-in NVIDIA® ConnectX-7 network link, two NVIDIA DGX Spark systems can be connected to support even larger models up to 405B parameters.

Delivery time
Delivery for the ASUS Ascent GX10 – NVIDIA DGX Spark typically takes 4–6 weeks from receipt of order. Please take this into account when planning your project.

Technical data

feature	specification
AI computing power	1 petaflops (FP4, sparse) (NVIDIA Grace Blackwell)
R.A.M.	128GB RAM LPDDR5x (Unified memory)
RAM bandwidth	273 GB/s (256-bit memory interface)
GPU	NVIDIA Blackwell Architecture
CPU	20 core ARM: 10 Cortex-X925 + 10 Cortex-A725 ARM
memory	1 x M.2 2242/2230 PCIe Gen5x4 (1TB/2TB/4TB Value & Performance), PCIe Gen4 compatible
Wi-Fi & Bluetooth	AW-EM637 Wi-Fi 7 (Gig+) 2x2 + Bluetooth® 5.3 with Bluetooth LE
network	1 x 10 GBit LAN 1 x ConnectX CX-7 Smart NIC (200 Gbps, 2 x QSFP)
USB ports	3 x USB 3.2 Gen 2x2 Type-C, 20Gbps, alternate mode (DisplayPort 2.1) 1 x USB 3.2 Gen 2x2 Type-C, 20 Gbps, with PD in(180W EPR PD3.1 SPEC)
Additional ports & features	1 x HDMI 2.1 1 x Kensington Lock 1 x on/off button
Operating system	NVIDIA DGX™ OS (Ubuntu Linux)
Dimensions	150 x 150 x 51 mm
Weight	1.48 kg

Scope of delivery

1 × ASUS Ascent GX 10 (based on NVIDIA DGX Spark)
power supply
Power cable (EU)
Note: QSFP cable for the ConnectX CX-7 network interface, for connecting two NVIDIA DGX Spark systems, is not included - it is sold separately

Areas of application

Industry & Research : Development and fine-tuning of AI LLM models with up to 200B parameters (or 405B when using two ASUS Ascent GX10 / DGX Sparks)
Universities & Colleges: Experiments and hands-on experience for students at large LLM models
Offices, business environment: Running Agentic AI, especially for data protection - e.g. for processing customer inquiries, offline email processing, RAGs with sensitive company data.

NVIDIA DGX Spark product variants

The NVIDIA DGX Spark platform is built on NVIDIA components at its core. Every variant features the Grace Blackwell GB10 superchip and 128 GB of RAM with 273 GB/s memory bandwidth. Memory bandwidth has the greatest impact on token/s throughput rates.

The operating system (NVIDIA DGX OS, based on Ubuntu) and the NVIDIA AI software stack are the same in all cases.

Various providers build a housing around it and sell the device under their own brand name:

NVIDIA directly under "NVIDIA DGX Spark" (formerly Project Digits)
ASUS as "ASUS Ascent GX10"
HP as "HP ZGX Nano AI Station G1n"
Dell as "Dell Pro Max With GB10"
Lenovo as "Lenovo ThinkStation PGX"

In our opinion, these devices are largely comparable in terms of AI performance. We recommend customers purchase the brand of their choice and pay particular attention to other factors such as service provided by the respective brand.

As an ASUS IoT partner, we offer the ASUS Ascent GX10 variant of the NVIDIA DGX Spark in our shop. ASUS's strength lies in developing particularly optimized cooling concepts, which they have also implemented with the DGX Spark in the form of the ASUS Ascent GX10.

Software stack / supported software

NVIDIA DGX OS (operating system based on Ubuntu)
NVIDIA NIM (Microservices as building blocks for AI solutions)
NVIDIA AI Blueprints (example applications, e.g. RAG workflows)
PyTorch (framework for machine learning and deep neural networks)
TensorFlow ( Framework for Machine Learning and Deep Neural Networks )
NVIDIA Riva (speech recognition, translation, and speech output)
NVIDIA Holoscan (platform for sensor data processing)
NVIDIA Metropolis (Vision AI Application Platform)
NVIDIA Isaac (platform for developing robots, e.g. autonomous vehicles)
CUDA (toolkit for accelerating and parallelizing computations on GPUs)
cuDNN (Acceleration of Deep Neural Networks on GPUs)
cuBLAS (Linear Algebra Frameworks for Accelerating AI and High Performance Computing (HPC) Applications)
TensorRT (SDK for high-performance deep learning inferencing)
NCCL (NVIDIA Collective Communications library, for distributing computing power across multiple nodes)
vLLM + ollama + llama.cpp - Large Language Model Inferencing Frameworks