Ultra-Small AI Supercomputer

ASUS Ascent GX10 — Based on NVIDIA DGX™ Spark

The ASUS Ascent GX10, accelerated by the NVIDIA GB10 Grace Blackwell Superchip and the NVIDIA AI software stack, provides a full-stack solution for AI development and deployment. Its compact design facilitates seamless integration and deployment, enabling powerful AI performance for innovators who demand excellence. With advanced AI tools and NVIDIA® ConnectX®-7, this small-scale server enhances your AI capabilities while empowering your unique solutions.

ASUS Ascent GX10 on desk showing 1 petaFLOP highlight
NVIDIA® GB10
Grace Blackwell Superchip
NVIDIA® Blackwell GPU
128 GB Unified Memory
1 petaFLOP
AI performance (FP4)
NVLink™-C2C
CPU↔GPU unified memory
NVIDIA® ConnectX®-7
Dual-GX10 stacking
DGX™ OS (Ubuntu)
NVIDIA AI Software Stack
Design

Revolutionary AI Performance on Your Desktop

The groundbreaking ASUS Ascent GX10 AI Supercomputer, powered by the state-of-the-art NVIDIA® GB10 Grace Blackwell Superchip found in the NVIDIA DGX Spark, brings petaflop-scale AI computing capabilities directly to the desks of developers, AI researchers, and data scientists. This innovative device is designed to empower local AI development with its exceptional performance and advanced features.

Unparalleled AI Performance

Unparalleled AI Performance

  • Up to 1 petaFLOP of AI performance using FP4: Delivers up to 1 petaFLOP of AI performance to power large AI workloads.
  • 128 GB LPDDR5x Coherent Unified System Memory: Empowers model development, experimentation, and inferencing with ample memory capacity.

Cutting-Edge Architecture

  • NVIDIA GB10 Grace Blackwell Superchip: Central to the ASUS Ascent GX10, this advanced chip features a robust Blackwell GPU with fifth-generation Tensor Cores and FP4 support.
  • High-Performance 20-Core Arm CPU: Enhances data preprocessing and orchestration, accelerating model tuning and real-time inferencing.
  • NVLink™-C2C Technology: Provides a cohesive CPU+GPU memory model with five times the bandwidth of PCIe 5.0.
Cutting-Edge Architecture
Handles Large Parameter Gen AI Models

Handles Large Parameter Gen AI Models

  • Support for AI Models 200 Billion Parameters: Prototype, fine-tune, and infer the latest AI reasoning models directly on your desktop.​
  • Integrated NVIDIA® ConnectX®-7 Network Technology: Link two ASUS Ascent GX10 systems to handle even larger models, such as Llama 3.1 with 405 billion parameters.
AI Networking

Meets Next-Gen Connectivity: NVIDIA® ConnectX-7

The ASUS Ascent GX10 integrates NVIDIA® ConnectX-7 to deliver ultra-high-speed networking, enabling rapid data transfer and low-latency communication across distributed AI workloads.

Data throughput

Ultra-Fast AI Data Throughput

Ultra-fast bandwidth ensures rapid data transfer between nodes, perfect for large-scale distributed AI workloads.

Security

Secure, Intelligent Networking

Built-in hardware acceleration for TLS, IPsec, and MACsec ensures encrypted data transmission without CPU overhead.

Precision timing

Precision-Critical Performance

IEEE 1588v2 PTP support enables microsecond-level time synchronization for time-sensitive AI and edge computing applications.

AI Experience

Integrated AI Software Stack for Seamless Development

Preloaded AI for Instant Development

Preloaded AI for Instant Development

NVIDIA DGX OS (Ubuntu-based) – Optimized AI environment, ready to use. NVIDIA AI Software Stack – Preloaded frameworks, SDKs, and tools for fast deployment.

Optimized AI Tools & AI Frameworks

Optimized AI Tools & AI Frameworks

CUDA, PyTorch, TensorFlow, Jupyter – Optimized for AI model development & inference. NVIDIA TensorRT – High-performance AI inference engine. NVIDIA NIMs & Blueprints – Prebuilt AI workflows & microservices.

Industry-Leading AI Model Support

Industry-Leading Model Support

DeepSeek R1 – AI inference optimized up to 70B parameters. Llama 3.1 – Generative AI up to 405B parameters (dual-GX10). Meta, Google models – Broad compatibility with industry-leading AI frameworks.

Exploded cooling view of ASUS Ascent GX10

Thermal

Precision-Crafted for Ultimate Thermal Efficiency

Tackle AI's most demanding workloads with the ASUS Ascent GX10. Its advanced dual-fan design with 7-level control delivers smooth, precise airflow. With 1.6× more efficient thermal coverage than comparable compact systems, the GX10 stays cooler and consistently performs at its peak.

Scalable

Engineered for Maximum Efficiency

  • Optimized cooling design ensures sustained AI performance under heavy workloads.
  • Compact form factor delivering high-density AI computing in a small footprint.
Scalable ASUS Ascent GX10

Connectivity

I/O Ports

Silver ASUS GX10 AI supercomputer with rear ports on dark blue background. Silver ASUS GX10 AI supercomputer with rear ports on dark blue background.

1Kensington Lock Slot

2 1 × USB 3.2 Gen 2×2 Type-C, 20 Gbps,
alternate mode (DisplayPort 2.1), with PD in (180 W EPR PD3.1 SPEC)

3 3 × USB 3.2 Gen 2×2 Type-C, 20 Gbps,
alternate mode (DisplayPort 2.1)

4HDMI 2.1b Port

510 GbE LAN

6NVIDIA ConnectX-7 NIC

Application

Local Development, Scalable Deployment

Local Development, Scalable Deployment

Seamless Transition to Cloud

Leverage NVIDIA AI platform software architecture to move models from desktop environments to DGX Cloud or any accelerated cloud or data-center infrastructure with minimal code adjustments.

Cost-Effective Experimentation Platform

Free up essential compute resources in clusters better suited for training and deploying production models.

Prototyping
Prototyping
Fine Tuning / Inference
Fine Tuning / Inference
Data Science
Data Science

Get Started

Sign up to be notified when ASUS Ascent GX10 becomes available.

Specifications and product images are subject to change.

FAQ

AI applications often require a bigger memory. With the NVIDIA Blackwell GPU that supports 128GB of unified memory, ASUS Ascent GX10 is an AI supercomputer that enables faster training, better real-time inference, and support larger models like LLMs.
The Ascent GX10 comes pre-loaded with NVIDIA AI software stack, which developers can use frameworks such as PyTorch, TensorFlow, JAX, Ollama. The NVIDIA AI Software Stack also allows users to run open source model on the system.
For larger LLM fine-tuning or high performing simulation and scientific computing, sometimes you might need to scale the workload of your Ascent GX10 unit. A maximum of two GX10 can be clustered together to scale the workloads beyond a single unit. Employ the integrated NVIDIA ConnectX-7 Network Technology to enable clustering.
The ASUS Ascent GX10 showcases meticulous engineering with a thermal design built to handle demanding workloads effortlessly. Every element of its cooling system reflects careful craftsmanship, ensuring sustained high performance in an impressively compact form.
The Ascent GX10 supports a wide range of AI workloads. In generative AI, it enables fine-tuning and running large language models (LLMs) for advanced chatbots and virtual assistants. It also excels in computer vision tasks such as object detection and tracking, as well as predictive analytics for risk assessment and demand forecasting. Finally, the GX10 is well-suited for simulation and research, making it a versatile AI computing solution for multiple industries.
With up to 1 petaFLOP of AI performance, ASUS Ascent GX10 provides institutions, educators, and students the chance to experience hands-on AI training, model fine-tuning, and deploying large scale AI models. Users can now prototype and experiment with models right on campus, instead of solely depdent on cloud resources.
Ascent GX10 is a compact AI supercomputer that removes the barrier to high-performnace AI computing by enabling local AI workloads and avoiding recurring cloud computing costs, all while keeping data confidential. With the compact 1.6L chassis, it's designed to fit into standard office environments. Whether you're in manufacturing, retail, finance, healthcare, or technology, Ascent GX10 enables faster insights and efficient operations.