{"product_id":"supermicro-as-8126gs-nb3rt-01-g2-ai-gpu-server-dual-amd-epyc-9575f-8x-nvidia-hgx-b300-gpus-3tb-ddr5-800gbe-networking-w647x","title":"Supermicro AS-8126GS-NB3RT-01-G2 AI GPU Server | Dual AMD EPYC 9575F | 8x NVIDIA HGX B300 GPUs | 3TB DDR5 | 800GbE Networking","description":"\u003ch1\u003e\u003cstrong\u003eSupermicro AS-8126GS-NB3RT-01-G2 AI GPU Server\u003c\/strong\u003e\u003c\/h1\u003e\n\u003ch2\u003e\u003cstrong\u003e8x NVIDIA HGX B300 GPUs with Dual AMD EPYC 9575F Processors for AI Training and HPC\u003c\/strong\u003e\u003c\/h2\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cspan style=\"color: rgb(255, 42, 0);\"\u003e\u003cstrong\u003eNote: New Supermicro product with a 3-year warranty. Available exclusively for U.S. end users. Limited quantities are available for immediate shipment. \u003c\/strong\u003e\u003c\/span\u003e\u003c\/strong\u003e\u003cstrong\u003e\u003cspan style=\"color: rgb(255, 42, 0);\"\u003e\u003cstrong\u003eThis product is export restricted and may not be relocated or transferred without compliance with all applicable regulations.\u003c\/strong\u003e\u003c\/span\u003e\u003c\/strong\u003e\u003c\/p\u003e\n\u003cp\u003eThe \u003cstrong\u003eSupermicro AS-8126GS-NB3RT-01-G2\u003c\/strong\u003e is a next-generation AI GPU server purpose-built for large-scale artificial intelligence, machine learning, generative AI, and high-performance computing workloads. Featuring \u003cstrong\u003e8 NVIDIA HGX B300 GPUs\u003c\/strong\u003e, dual \u003cstrong\u003eAMD EPYC 9575F processors\u003c\/strong\u003e, and \u003cstrong\u003e3TB DDR5-6400 memory\u003c\/strong\u003e, this platform delivers exceptional compute density for enterprise AI factories, research institutions, and cloud-scale infrastructure.\u003c\/p\u003e\n\u003cp\u003ePowered by \u003cstrong\u003e256 total AMD EPYC cores\u003c\/strong\u003e running at 3.3GHz, the server provides the processing performance required to support large language model (LLM) training, deep learning frameworks, scientific simulations, and advanced analytics. The integrated \u003cstrong\u003eNVIDIA HGX B300 platform\u003c\/strong\u003e leverages NVIDIA Blackwell architecture and high-speed NVLink connectivity to accelerate AI model training and inference at scale.\u003c\/p\u003e\n\u003cp\u003eFor data-intensive workloads, the system includes \u003cstrong\u003e8× 7.6TB E1.S NVMe SSDs\u003c\/strong\u003e, providing \u003cstrong\u003e60.8TB of high-performance storage\u003c\/strong\u003e, alongside \u003cstrong\u003e2× 1.9TB M.2 NVMe SSDs\u003c\/strong\u003e for operating system and application deployment. Ultra-fast networking is delivered through an \u003cstrong\u003e8-port 800GbE\/XDR800 fabric\u003c\/strong\u003e and \u003cstrong\u003e4 additional 200GbE\/NDR200 ports\u003c\/strong\u003e, enabling seamless communication across AI clusters and supercomputing environments.\u003c\/p\u003e\n\u003cp\u003eDesigned for continuous operation in enterprise data centers, the AS-8126GS-NB3RT-01-G2 incorporates \u003cstrong\u003eredundant 6600W Titanium-level power supplies\u003c\/strong\u003e, advanced cooling, ECC memory protection, and enterprise-grade management capabilities. The platform supports leading AI software frameworks including NVIDIA CUDA, TensorRT, NCCL, RAPIDS, and NVIDIA AI Enterprise.\u003c\/p\u003e\n\u003ch3\u003e\u003cstrong\u003eProduct Specifications – Supermicro AS-8126GS-NB3RT-01-G2 AI GPU Server\u003cbr\u003e\u003ca href=\"https:\/\/www.supermicro.com\/en\/products\/system\/datasheet\/as%20-8126gs-nb3rt-01-g2?_gl=1*57y6kn*_up*MQ..*_ga*Mzg3Njk5MDg5LjE3ODA0ODg5MzQ.*_ga_BMMP56ELKC*czE3ODA0ODg5MzMkbzEkZzAkdDE3ODA0ODg5MzMkajYwJGwwJGgw*_ga_K1TNNPLV4M*czE3ODA0ODg5MzMkbzEkZzAkdDE3ODA0ODg5MzMkajYwJGwwJGgxNTA5MjI1OTg0\" title=\"Supermicro AS-8126GS-NB3RT-01-G2 - Data Sheet\" rel=\"noopener\" target=\"_blank\"\u003eCLICK HERE FOR DATA SHEET\u003c\/a\u003e\u003cbr\u003e\u003c\/strong\u003e\u003c\/h3\u003e\n\u003ctable\u003e\n\u003cthead\u003e\n\u003ctr\u003e\n\u003cth\u003e\u003cstrong\u003eSpecification Category\u003c\/strong\u003e\u003c\/th\u003e\n\u003cth\u003e\u003cstrong\u003eDetails\u003c\/strong\u003e\u003c\/th\u003e\n\u003c\/tr\u003e\n\u003c\/thead\u003e\n\u003ctbody\u003e\n\u003ctr\u003e\n\u003ctd\u003e\u003cstrong\u003eProduct Name\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd\u003eSupermicro AS-8126GS-NB3RT-01-G2\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003e\u003cstrong\u003eManufacturer\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd\u003eSupermicro\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003e\u003cstrong\u003eProduct Type\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd\u003eAI Training \u0026amp; High-Performance Computing (HPC) GPU Server\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003e\u003cstrong\u003eForm Factor\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd\u003e8U Rackmount GPU Server\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003e\u003cstrong\u003eSystem Architecture\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd\u003eNVIDIA HGX B300 Platform\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003e\u003cstrong\u003eProcessor Configuration\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd\u003eDual-Socket Server\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003e\u003cstrong\u003eCPU Model\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd\u003e2 × AMD EPYC™ 9575F Processors\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003e\u003cstrong\u003eCPU Cores\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd\u003e128 Cores per CPU (256 Total Cores)\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003e\u003cstrong\u003eCPU Base Frequency\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd\u003e3.3 GHz\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003e\u003cstrong\u003eProcessor Architecture\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd\u003eAMD EPYC 9005 Series (Turin)\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003e\u003cstrong\u003eGPU Configuration\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd\u003e8 × NVIDIA HGX™ B300 GPUs\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003e\u003cstrong\u003eGPU Architecture\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd\u003eNVIDIA Blackwell Architecture\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003e\u003cstrong\u003eGPU Interconnect\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd\u003eNVIDIA NVLink High-Speed GPU Fabric\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003e\u003cstrong\u003eSystem Memory\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd\u003e3 TB DDR5-6400 ECC Registered Memory\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003e\u003cstrong\u003eMemory Technology\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd\u003eDDR5 RDIMM\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003e\u003cstrong\u003eMemory Speed\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd\u003e6400 MT\/s\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003e\u003cstrong\u003eMemory Protection\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd\u003eECC (Error-Correcting Code)\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003e\u003cstrong\u003ePrimary Storage\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd\u003e8 × 7.6TB E1.S NVMe SSDs\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003e\u003cstrong\u003eStorage Capacity\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd\u003e60.8TB Raw NVMe Storage\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003e\u003cstrong\u003eStorage Interface\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd\u003ePCIe NVMe\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003e\u003cstrong\u003eBoot Drives\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd\u003e2 × 1.9TB M.2 NVMe SSDs\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003e\u003cstrong\u003eM.2 Storage Capacity\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd\u003e3.8TB Total\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003e\u003cstrong\u003eNetworking Fabric\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd\u003e8-Port 800GbE \/ XDR800 High-Speed Interconnect\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003e\u003cstrong\u003eNetwork Speed\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd\u003eUp to 800Gbps per port\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003e\u003cstrong\u003eAdditional Networking\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd\u003e2 × Dual-Port 200GbE \/ NDR200 Network Adapters\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003e\u003cstrong\u003eAdditional Network Ports\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd\u003e4 × 200GbE Ports\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003e\u003cstrong\u003eCluster Connectivity\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd\u003eOptimized for AI Factories and Multi-Node GPU Clusters\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003e\u003cstrong\u003ePower Supply Configuration\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd\u003eRedundant Power Supplies\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003e\u003cstrong\u003ePower Rating\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd\u003e6600W Titanium-Level Efficiency\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003e\u003cstrong\u003ePower Redundancy\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd\u003eYes\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003e\u003cstrong\u003eCooling System\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd\u003eHigh-Performance Redundant Cooling Architecture\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003e\u003cstrong\u003eRemote Management\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd\u003eEnterprise Remote Server Management Support\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003e\u003cstrong\u003eSecurity Features\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd\u003eSecure Boot, TPM Support, Hardware Root of Trust\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003e\u003cstrong\u003eExpansion Architecture\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd\u003eNVIDIA HGX GPU Baseboard with High-Speed PCIe Infrastructure\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003e\u003cstrong\u003eAI Framework Support\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd\u003eNVIDIA CUDA, TensorRT, NCCL, RAPIDS, NVIDIA AI Enterprise\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003e\u003cstrong\u003eOperating System Support\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd\u003eEnterprise Linux Distributions and HPC Software Stacks\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003e\u003cstrong\u003eDeployment Environment\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd\u003eAI Data Centers, Research Labs, Cloud AI Infrastructure\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003e\u003cstrong\u003eTarget Workloads\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd\u003eLarge Language Models (LLMs), Generative AI, Deep Learning, AI Training, AI Inference, HPC, Scientific Simulation\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003e\u003cstrong\u003eReliability Features\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd\u003eRedundant Power, ECC Memory, Enterprise-Class Components\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003e\u003cstrong\u003eStorage Use Cases\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd\u003eAI Datasets, Model Checkpoints, High-Speed Data Pipelines, Large-Scale Analytics\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003e\u003cstrong\u003eNetwork Use Cases\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd\u003eDistributed AI Training, Multi-Node Clusters, AI Supercomputing, HPC Fabrics\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003c\/tbody\u003e\n\u003c\/table\u003e\n\u003ch3\u003e\u003cstrong\u003eFAQs\u003c\/strong\u003e\u003c\/h3\u003e\n\u003cp\u003e\u003cstrong\u003e1. How many GPUs does the AS-8126GS-NB3RT-01-G2 include?\u003c\/strong\u003e\u003c\/p\u003e\n\u003cp\u003eIt features 8 NVIDIA HGX B300 GPUs for AI training, inference, and HPC workloads.\u003c\/p\u003e\n\u003cp\u003e\u003cstrong\u003e2. Which processors power this server?\u003c\/strong\u003e\u003c\/p\u003e\n\u003cp\u003eThe system is equipped with 2 AMD EPYC 9575F processors, providing 256 total CPU cores.\u003c\/p\u003e\n\u003cp\u003e\u003cstrong\u003e3. How much memory is installed?\u003c\/strong\u003e\u003c\/p\u003e\n\u003cp\u003eThe server includes 3TB DDR5-6400 ECC memory for memory-intensive AI and data analytics workloads.\u003c\/p\u003e\n\u003cp\u003e\u003cstrong\u003e4. What storage configuration does it offer?\u003c\/strong\u003e\u003c\/p\u003e\n\u003cp\u003eIt comes with 8× 7.6TB E1.S NVMe SSDs and 2× 1.9TB M.2 NVMe SSDs for high-speed storage and boot functionality.\u003c\/p\u003e\n\u003cp\u003e\u003cstrong\u003e5. What workloads is this server designed for?\u003c\/strong\u003e\u003c\/p\u003e\n\u003cp\u003eIt is optimized for LLM training, generative AI, machine learning, deep learning, AI inference, and high-performance computing (HPC).\u003c\/p\u003e","brand":"Super Micro","offers":[{"title":"Default Title","offer_id":48144063103137,"sku":"AS-8126GS-NB3RT-01-G2","price":17996790.0,"currency_code":"BRL","in_stock":true}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/2216\/8673\/files\/Supermicro-AS-8126GS-NB3RT-01-G2-AI-GPU-Server-Dual-AMD-EPYC-9575F-8x-NVIDIA-HGX-B300-GPUs-3TB-DDR5-800GbE-Networking_f1d5b819.webp?v=1792477401","url":"https:\/\/www.foxti.com.br\/products\/supermicro-as-8126gs-nb3rt-01-g2-ai-gpu-server-dual-amd-epyc-9575f-8x-nvidia-hgx-b300-gpus-3tb-ddr5-800gbe-networking-w647x","provider":"FoxTI","version":"1.0","type":"link"}