ASUS Ascent GX10 Barebone ARM v9.2-A CPU GB10 NVIDIA Blackwell GPU GB10 128GBLPDDR5x 1TBSSD PCIe Gen5x4

Ultra-small AI supercomputer
Based on NVIDIA DGX™ Spark
The ASUS Ascent GX10, accelerated by the NVIDIA GB10 Grace Blackwell superchip and the NVIDIA AI software stack, offers a complete solution for AI development and deployment. Its compact design facilitates seamless integration and deployment, enabling powerful AI performance for innovators who demand peak performance. With advanced AI tools and NVIDIA® ConnectX® 7, this small-scale server expands your AI capabilities and supports your unique solutions.

design
Revolutionary AI performance on your desktop
The groundbreaking ASUS Ascent GX10 AI supercomputer, equipped with the state-of-the-art NVIDIA® GB10 Grace Blackwell superchip from the NVIDIA DGX Spark, brings petaflop-level AI computing power directly to the desktops of developers, AI researchers, and data scientists. This innovative device is designed to support local AI development with its exceptional performance and advanced features.

Compact, powerful and scalable

Unrivaled AI performance
Up to 1 petaFLOP AI performance with FP4
Delivers up to 1 petaFLOP of AI performance for large AI workloads.

128GB LPDDR5x Coherent Unified System Memory
Enables model development, experimentation and inference with ample storage capacity.

ultramodern architecture
NVIDIA GB10 Grace Blackwell Superchip:
This advanced chip is the heart of the ASUS Ascent GX10 and features a robust Blackwell GPU with fifth generation Tensor Cores and FP4 support.

Powerful 20-core Arm CPU:
Improves data preprocessing and orchestration, accelerates model tuning and real-time inference.

NVLink™-C2C Technology:
Offers a coherent CPU+GPU memory model with five times the bandwidth of PCIe 5.0.

Handles AI models with large parameters
Support for AI models with 200 billion parameters
Develop prototypes, fine-tune them, and derive the latest AI conclusions directly on your desktop.

Integrated NVIDIA® ConnectX® 7 networking technology
Connect two ASUS Ascent GX10 systems to handle even larger models like Llama 3.1 with 405 billion parameters.

AI Networking
Meets next-generation connectivity: NVIDIA® ConnectX-7
The ASUS Ascent GX10 is equipped with NVIDIA® ConnectX-7 and offers an ultra-high-speed network that enables fast data transfer and low-latency communication for distributed AI workloads.

Ultra-fast AI data throughput
The bandwidth of up to 200 Gb/s enables lightning-fast data transfer between nodes - ideal for large, distributed AI workloads.

Secure, intelligent networking
The integrated hardware acceleration for TLS, IPsec and MACsec ensures encrypted data transmission without CPU overhead.

Precision-critical performance
Support for IEEE 1588v2 PTP enables microsecond-level time synchronization for time-critical AI and edge computing applications.

AI Experience
Integrated AI software stack for seamless development

Pre-installed AI for immediate development
NVIDIA DGX OS (Ubuntu-based) - Optimized AI environment, ready to use immediately.
NVIDIA AI Software Stack - Pre-installed frameworks, SDKs and tools for quick deployment.

Optimized AI tools & AI frameworks
CUDA, PyTorch, TensorFlow, Jupyter - Optimized for the development of AI models & inference.
NVIDIA TensorRT - Powerful AI inference engine.
NVIDIA NIMs & Blueprints - Pre-built AI workflows & microservices.

Industry-leading AI model support
DeepSeek R1 - Optimized AI inference with up to 70 parameters.
Llama 3.1 - Generative AI with up to 405B parameters (Dual-GX10).
Meta, Google Models - Broad compatibility with industry-leading AI frameworks.

Scalable
Designed for maximum efficiency

Optimized cooling design for sustained AI performance under heavy workloads
Compact factor that enables high-density AI computing in a small space

Application
Local development, scalable deployment

Seamless transition to the cloud
Leverage the software architecture of the NVIDIA AI platform to move models from desktop environments to the DGX Cloud or any other accelerated cloud or data center infrastructure with minimal code modifications.

Cost-efficient experimentation platform
Free up critical computing resources in clusters that are better suited for training and deploying production models.