ASUS Ascent GX10 AI Supercomputer (1TB)

ASUS Ascent GX10 AI Supercomputer (1TB)

Brand: ASUS
SKU: 90MS0371-M00040
13800.00 AED In stock Buy at Merchant

ASUS Ascent GX10 AI Supercomputer ASUS Ascent GX10 is a compact desktop AI supercomputer built for developers, AI researchers, data scientists and machine learning teams that need serious local compute for model prototyping, fine-tuning, inference and agentic AI workflows. Powered by the NVIDIA GB10 Grace Blackwell Superchip with an integrated NVIDIA Blackwell GPU, 128GB unified memory and up to 1000 AI TOPS, GX10 brings high-density AI performance into a small form factor for labs, offices and edge environments. Key Performance Highlights 1000 AI TOPS GX10 delivers up to 1000 AI TOPS for local AI development, inference processing and model experimentation. 128GB unified memory 128GB LPDDR5x unified system memory helps support large AI models, complex datasets and 200B model fine-tuning workflows referenced in the ASUS datasheet. High-speed networking Built-in 10G LAN, Wi-Fi 7 and NVIDIA ConnectX-7 connectivity help teams move data quickly and scale into more advanced workflows. Why Choose ASUS Ascent GX10? Extreme AI performance NVIDIA GB10 combines a Grace CPU and Blackwell GPU architecture to deliver compact, high-performance AI computing for serious development teams. Developer-optimised platform GX10 is built for AI researchers, developers, data scientists and machine learning teams that need a reliable local platform for experimentation and deployment. Scalable architecture NVIDIA NVLink-C2C supports ultra-fast CPU-GPU memory communication, while ConnectX-7 networking supports dual GX10 system stacking. Compact footprint At just 150 x 150 x 51 mm, GX10 fits into AI labs, offices, classrooms, research spaces and edge environments. Advanced thermal design The compact chassis is engineered for sustained AI performance and dependable operation in a small desktop form factor. Local AI control Run sensitive experiments, model tests and inference workloads closer to your team, with less dependency on remote compute for every iteration. AI Workloads and Use Cases AI model prototyping Build, test and refine AI models locally before moving workloads into production or cloud environments. Fine-tuning and inference Use GX10 for model fine-tuning, private inference, experimentation and repeated development cycles. Agentic AI workflows GX10 is a strong local platform for agentic AI systems that plan, execute tasks and manage workflows with reduced manual intervention. OpenClaw PC projects For teams searching for an OpenClaw PC, ASUS Ascent GX10 provides NVIDIA GB10 compute, Blackwell GPU acceleration, 128GB unified memory and fast networking. Computer vision and robotics Develop object detection, robotics simulation, industrial automation and edge AI workflows on a compact local system. Data science and simulation Run predictive analytics, risk modelling, scientific simulation and data-heavy AI experiments closer to your team. Ideal Users and Industries Ideal users AI researchers, AI developers, data scientists, machine learning engineers, computer vision teams, robotics engineers, IT architects and AI startup founders. Research and education A compact platform for universities, academic labs and research institutions working on deep learning, AI experiments and model development. Healthcare and biotech Useful for medical imaging, genomics, drug discovery, molecular modelling and healthcare AI development workflows. Finance and banking Supports predictive analytics, fraud detection, risk simulations and data science projects where local compute is valuable. Industrial AI and automation Suitable for robotics, manufacturing, automation, edge AI and smart industrial workflows. Media and content creation Can support AI-assisted video editing, rendering, creative workflows and local experimentation with generative AI tools. ASUS Ascent GX10 Specifications Processor and graphics NVIDIA GB10 with ARM v9.2-A CPU and integrated NVIDIA Blackwell GPU. Operating system NVIDIA DGX Base OS, Ubuntu Linux. Memory 128GB LPDDR5x unified system memory. Storage 1 x M.2 2242/2230 PCIe Gen5 x4, PCIe Gen4 compatible; 1TB, 2TB and 4TB value/performance options. Wireless and LAN AW-EM637 Wi-Fi 7, 2x2, Bluetooth 5 and 1 x 10G LAN. Rear I/O 3 x USB 3.2 Gen 2x2 Type-C, 1 x USB 3.2 Gen 2x2 Type-C with PD in, 1 x HDMI 2.1, 1 x ConnectX CX-7, 1 x 10G LAN and 1 x Kensington lock. Dimensions and weight 150 x 150 x 51 mm, 5.91 x 5.91 x 2.01 inches, approximately 1.48 kg. Box contents ASUS Ascent GX10, power cord, QSFP cable and inserts. Bundled accessories may vary by region and SKU. Specifications, content, accessories and product availability may vary by country and SKU. Confirm the exact configuration before ordering. ASUS Ascent GX10 FAQ What is ASUS Ascent GX10? ASUS Ascent GX10 is a compact desktop AI supercomputer powered by NVIDIA GB10 Grace Blackwell architecture. It is designed for local AI development, inference, fine-tuning, data science and edge AI workloads. Who should use the ASUS Ascent GX10? GX10 is suitable for AI researchers, developers, data scientists, machine learning engineers, robotics teams, computer vision teams, AI startups and organisations building local AI workflows. Can ASUS Ascent GX10 run agentic AI and OpenClaw PC workflows? Yes. GX10 is positioned as a strong local AI platform for agentic AI and OpenClaw PC workflows because it combines NVIDIA GB10 compute, Blackwell GPU acceleration, 128GB unified memory and fast networking in a compact system. How much memory does ASUS Ascent GX10 include? The ASUS datasheet lists 128GB LPDDR5x unified system memory. What storage options are available for ASUS Ascent GX10? The GX10 datasheet lists M.2 PCIe storage with 1TB, 2TB and 4TB value/performance options, depending on SKU and availability. Does ASUS Ascent GX10 support high-speed networking? Yes. The datasheet lists Wi-Fi 7, Bluetooth 5, 10G LAN and NVIDIA ConnectX CX-7 connectivity.

Variants (1)
  • Default Title — 13800.00 AED — In stock

AI Readiness

Good foundation, but some important product data is still missing.

78%