{"title":"AI\/HPC Shop","description":"\u003ch2\u003eHardware for AI, Machine Learning and Serious Computing\u003c\/h2\u003e\n\u003cp\u003eWelcome to the LIKUA AI\/HPC Shop. We have put together a focused selection of workstations, professional GPUs, server processors and AI systems for real development, engineering and scientific workloads.\u003c\/p\u003e\n\u003cp\u003eOur aim is not to list every piece of hardware available. We want to help AI developers, engineers, researchers and technical teams find capable, realistically available systems and components more easily.\u003c\/p\u003e\n\u003cp\u003eChoose the category that matches your workflow. Each category page explains how the hardware can be used for AI, machine learning, HPC and engineering applications.\u003c\/p\u003e","products":[{"product_id":"hp-zbook-fury-16-g1i-core-ultra-9-285hx-rtx-pro-4000-16-gb-64-gb-2-tb","title":"HP ZBook Fury 16 G1i – Core Ultra 9 285HX – RTX PRO 4000 16 GB – 64 GB – 2 TB","description":"\u003cp\u003eHigh-performance mobile workstation for AI, machine-learning and engineering development on the move. The NVIDIA GPU enables CUDA-accelerated workflows directly on the notebook.\u003c\/p\u003e\u003ch3\u003eTypical AI workloads\u003c\/h3\u003e\u003cul\u003e\n\u003cli\u003ePython, PyTorch and CUDA development\u003c\/li\u003e\n\u003cli\u003eLocal LLM inference and AI assistants\u003c\/li\u003e\n\u003cli\u003eComputer vision and image analysis\u003c\/li\u003e\n\u003cli\u003eRAG demos, data science and prototyping\u003c\/li\u003e\n\u003cli\u003eCAD\/CAE, rendering and technical visualization\u003c\/li\u003e\n\u003c\/ul\u003e\u003cp\u003e\u003cstrong\u003eGPU memory:\u003c\/strong\u003e 16 GB of GPU memory is well suited to development, computer vision and many small-to-medium local AI workloads. Model size and speed also depend on precision, quantization and the notebook's power profile.\u003c\/p\u003e\u003cp\u003e\u003cstrong\u003eBest suited to:\u003c\/strong\u003e developers and engineers who value mobility more than the maximum sustained performance and expandability of a tower workstation.\u003c\/p\u003e","brand":"HP","offers":[{"title":"Default Title","offer_id":46818057093222,"sku":"98L80ET#ABD","price":10190.0,"currency_code":"EUR","in_stock":false}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0680\/6572\/2470\/files\/hp-zbook-fury-16-g1i-98l80et-hero.jpg?v=1788262241"},{"product_id":"hp-zbook-fury-g1i-18-core-ultra-9-285hx-rtx-pro-3000-12-gb-64-gb-2-tb","title":"HP ZBook Fury G1i 18 – Core Ultra 9 285HX – RTX PRO 3000 12 GB – 64 GB – 2 TB","description":"\u003cp\u003eHigh-performance mobile workstation for AI, machine-learning and engineering development on the move. The NVIDIA GPU enables CUDA-accelerated workflows directly on the notebook.\u003c\/p\u003e\u003ch3\u003eTypical AI workloads\u003c\/h3\u003e\u003cul\u003e\n\u003cli\u003ePython, PyTorch and CUDA development\u003c\/li\u003e\n\u003cli\u003eLocal LLM inference and AI assistants\u003c\/li\u003e\n\u003cli\u003eComputer vision and image analysis\u003c\/li\u003e\n\u003cli\u003eRAG demos, data science and prototyping\u003c\/li\u003e\n\u003cli\u003eCAD\/CAE, rendering and technical visualization\u003c\/li\u003e\n\u003c\/ul\u003e\u003cp\u003e\u003cstrong\u003eGPU memory:\u003c\/strong\u003e 12 GB of GPU memory is well suited to development, computer vision and many small-to-medium local AI workloads. Model size and speed also depend on precision, quantization and the notebook's power profile.\u003c\/p\u003e\u003cp\u003e\u003cstrong\u003eBest suited to:\u003c\/strong\u003e developers and engineers who value mobility more than the maximum sustained performance and expandability of a tower workstation.\u003c\/p\u003e","brand":"HP","offers":[{"title":"Default Title","offer_id":46824891580518,"sku":"98L73ET#ABD","price":9460.0,"currency_code":"EUR","in_stock":false}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0680\/6572\/2470\/files\/hp-zbook-fury-g1i-18-98l73et-1.jpg?v=1788266738"},{"product_id":"lenovo-thinkpad-p16-gen-3-core-ultra-9-275hx-rtx-pro-4000-16-gb-64-gb-1-tb","title":"Lenovo ThinkPad P16 Gen 3 – Core Ultra 9 275HX – RTX PRO 4000 16 GB – 64 GB – 1 TB","description":"\u003cp\u003eHigh-performance mobile workstation for AI, machine-learning and engineering development on the move. The NVIDIA GPU enables CUDA-accelerated workflows directly on the notebook.\u003c\/p\u003e\u003ch3\u003eTypical AI workloads\u003c\/h3\u003e\u003cul\u003e\n\u003cli\u003ePython, PyTorch and CUDA development\u003c\/li\u003e\n\u003cli\u003eLocal LLM inference and AI assistants\u003c\/li\u003e\n\u003cli\u003eComputer vision and image analysis\u003c\/li\u003e\n\u003cli\u003eRAG demos, data science and prototyping\u003c\/li\u003e\n\u003cli\u003eCAD\/CAE, rendering and technical visualization\u003c\/li\u003e\n\u003c\/ul\u003e\u003cp\u003e\u003cstrong\u003eGPU memory:\u003c\/strong\u003e 16 GB of GPU memory is well suited to development, computer vision and many small-to-medium local AI workloads. Model size and speed also depend on precision, quantization and the notebook's power profile.\u003c\/p\u003e\u003cp\u003e\u003cstrong\u003eBest suited to:\u003c\/strong\u003e developers and engineers who value mobility more than the maximum sustained performance and expandability of a tower workstation.\u003c\/p\u003e","brand":"Lenovo","offers":[{"title":"Default Title","offer_id":46826884464742,"sku":"21RQ005XGE","price":7090.0,"currency_code":"EUR","in_stock":false}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0680\/6572\/2470\/files\/lenovo-thinkpad-p16-gen3-21rq005xge-hero.jpg?v=1788268159"},{"product_id":"hp-zbook-fury-16-g1i-core-ultra-9-285hx-rtx-pro-3000-12-gb-64-gb-1-tb","title":"HP ZBook Fury 16 G1i – Core Ultra 9 285HX – RTX PRO 3000 12 GB – 64 GB – 1 TB","description":"\u003cp\u003eHigh-performance mobile workstation for AI, machine-learning and engineering development on the move. The NVIDIA GPU enables CUDA-accelerated workflows directly on the notebook.\u003c\/p\u003e\u003ch3\u003eTypical AI workloads\u003c\/h3\u003e\u003cul\u003e\n\u003cli\u003ePython, PyTorch and CUDA development\u003c\/li\u003e\n\u003cli\u003eLocal LLM inference and AI assistants\u003c\/li\u003e\n\u003cli\u003eComputer vision and image analysis\u003c\/li\u003e\n\u003cli\u003eRAG demos, data science and prototyping\u003c\/li\u003e\n\u003cli\u003eCAD\/CAE, rendering and technical visualization\u003c\/li\u003e\n\u003c\/ul\u003e\u003cp\u003e\u003cstrong\u003eGPU memory:\u003c\/strong\u003e 12 GB of GPU memory is well suited to development, computer vision and many small-to-medium local AI workloads. Model size and speed also depend on precision, quantization and the notebook's power profile.\u003c\/p\u003e\u003cp\u003e\u003cstrong\u003eBest suited to:\u003c\/strong\u003e developers and engineers who value mobility more than the maximum sustained performance and expandability of a tower workstation.\u003c\/p\u003e","brand":"HP","offers":[{"title":"Default Title","offer_id":46828115165286,"sku":"98L79ET#ABD","price":8270.0,"currency_code":"EUR","in_stock":false}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0680\/6572\/2470\/files\/hp-zbook-fury-16-g1i-98l79et-hero.jpg?v=1788269012"},{"product_id":"lenovo-thinkpad-p16-gen-3-core-ultra-9-275hx-rtx-pro-3000-12-gb-64-gb-1-tb","title":"Lenovo ThinkPad P16 Gen 3 – Core Ultra 9 275HX – RTX PRO 3000 12 GB – 64 GB – 1 TB","description":"\u003cp\u003eHigh-performance mobile workstation for AI, machine-learning and engineering development on the move. The NVIDIA GPU enables CUDA-accelerated workflows directly on the notebook.\u003c\/p\u003e\u003ch3\u003eTypical AI workloads\u003c\/h3\u003e\u003cul\u003e\n\u003cli\u003ePython, PyTorch and CUDA development\u003c\/li\u003e\n\u003cli\u003eLocal LLM inference and AI assistants\u003c\/li\u003e\n\u003cli\u003eComputer vision and image analysis\u003c\/li\u003e\n\u003cli\u003eRAG demos, data science and prototyping\u003c\/li\u003e\n\u003cli\u003eCAD\/CAE, rendering and technical visualization\u003c\/li\u003e\n\u003c\/ul\u003e\u003cp\u003e\u003cstrong\u003eGPU memory:\u003c\/strong\u003e 12 GB of GPU memory is well suited to development, computer vision and many small-to-medium local AI workloads. Model size and speed also depend on precision, quantization and the notebook's power profile.\u003c\/p\u003e\u003cp\u003e\u003cstrong\u003eBest suited to:\u003c\/strong\u003e developers and engineers who value mobility more than the maximum sustained performance and expandability of a tower workstation.\u003c\/p\u003e","brand":"Lenovo","offers":[{"title":"Default Title","offer_id":46832684531814,"sku":"21RQ005UGE","price":6510.0,"currency_code":"EUR","in_stock":false}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0680\/6572\/2470\/files\/lenovo-thinkpad-p16-gen3-21rq005uge-hero.jpg?v=1788272145"},{"product_id":"nvidia-l40s-48-gb-data-center-gpu-ada-lovelace-pcie-4-0-350-w","title":"NVIDIA L40S 48 GB Data Center GPU – Ada Lovelace – PCIe 4.0 – 350 W","description":"\u003cp\u003eData-center accelerator for sustained AI\/HPC workloads in server environments with 48 GB of local accelerator memory.\u003c\/p\u003e\u003ch3\u003eAI \u0026amp; HPC use\u003c\/h3\u003e\u003cul\u003e\n\u003cli\u003eCUDA-based AI inference, training and generative AI\u003c\/li\u003e\n\u003cli\u003eBatch inference and computer vision\u003c\/li\u003e\n\u003cli\u003eMulti-user and server-based compute environments\u003c\/li\u003e\n\u003cli\u003eGPU-accelerated scientific computing and HPC\u003c\/li\u003e\n\u003cli\u003eSustained centralized workloads\u003c\/li\u003e\n\u003c\/ul\u003e\u003cp\u003eThe L40S combines substantial GPU memory with NVIDIA CUDA and professional graphics\/AI capabilities.\u003c\/p\u003e\u003cp\u003e\u003cstrong\u003eBest suited to:\u003c\/strong\u003e laboratories, engineering teams and companies operating centralized AI\/HPC infrastructure.\u003c\/p\u003e","brand":"NVIDIA","offers":[{"title":"Default Title","offer_id":46833871487078,"sku":"900-2G133-0080-000","price":10040.0,"currency_code":"EUR","in_stock":false}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0680\/6572\/2470\/files\/nvidia-l40s-900-2g133-0080-000-hero.jpg?v=1788272835"},{"product_id":"amd-instinct-mi210-64-gb-hbm2e-ai-hpc-accelerator-cdna-2-pcie-4-0","title":"AMD Instinct MI210 64 GB HBM2e AI\/HPC Accelerator – CDNA 2 – PCIe 4.0","description":"\u003cp\u003eData-center accelerator for sustained AI\/HPC workloads in server environments with 64 GB of local accelerator memory.\u003c\/p\u003e\u003ch3\u003eAI \u0026amp; HPC use\u003c\/h3\u003e\u003cul\u003e\n\u003cli\u003eROCm-based machine-learning and scientific-computing workloads\u003c\/li\u003e\n\u003cli\u003eBatch inference and computer vision\u003c\/li\u003e\n\u003cli\u003eMulti-user and server-based compute environments\u003c\/li\u003e\n\u003cli\u003eGPU-accelerated scientific computing and HPC\u003c\/li\u003e\n\u003c\/ul\u003e\u003cp\u003eThe MI210 is particularly relevant to HPC and supported AI frameworks in the AMD ROCm ecosystem.\u003c\/p\u003e\u003cp\u003e\u003cstrong\u003eBest suited to:\u003c\/strong\u003e laboratories, engineering teams and companies operating centralized AI\/HPC infrastructure.\u003c\/p\u003e","brand":"AMD","offers":[{"title":"Default Title","offer_id":46836545060966,"sku":"100-300000008H","price":9420.0,"currency_code":"EUR","in_stock":false}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0680\/6572\/2470\/files\/amd-instinct-mi210-product-view.webp?v=1788278792"},{"product_id":"nvidia-rtx-pro-4500-blackwell-32-gb-gddr7-ecc-pcie-5-0","title":"NVIDIA RTX PRO 4500 Blackwell – 32 GB GDDR7 ECC – PCIe 5.0","description":"\u003cp\u003eProfessional NVIDIA RTX PRO workstation GPU with substantial GB of graphics memory for local AI, machine learning, CUDA development and technical visualization.\u003c\/p\u003e\u003ch3\u003eAI \u0026amp; ML use\u003c\/h3\u003e\u003cul\u003e\n\u003cli\u003eLocal LLM inference and AI assistants\u003c\/li\u003e\n\u003cli\u003eFine-tuning selected models\u003c\/li\u003e\n\u003cli\u003eComputer vision and deep learning\u003c\/li\u003e\n\u003cli\u003eDiffusion models and generative AI\u003c\/li\u003e\n\u003cli\u003ePyTorch, TensorFlow and CUDA development\u003c\/li\u003e\n\u003cli\u003eCAD\/CAE, rendering and visualization\u003c\/li\u003e\n\u003c\/ul\u003e\u003ch3\u003eWhy GPU memory matters\u003c\/h3\u003e\u003cp\u003e24 GB of VRAM is a practical capacity for many local LLM, computer-vision and generative-AI workflows. Whether a specific model fits also depends on precision, quantization, context length and batch size.\u003c\/p\u003e\u003cp\u003e\u003cstrong\u003eBest suited to:\u003c\/strong\u003e professional local AI development and combined AI\/engineering workflows.\u003c\/p\u003e","brand":"NVIDIA","offers":[{"title":"Default Title","offer_id":46837615001702,"sku":"900-5G147-2250-000","price":5040.0,"currency_code":"EUR","in_stock":false}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0680\/6572\/2470\/files\/nvidia-rtx-pro-4500-blackwell-product.jpg?v=1788276919"},{"product_id":"nvidia-rtx-pro-4000-sff-blackwell-24-gb-gddr7-ecc-70-w","title":"NVIDIA RTX PRO 4000 SFF Blackwell – 24 GB GDDR7 ECC – 70 W","description":"\u003cp\u003eProfessional NVIDIA RTX PRO workstation GPU with substantial GB of graphics memory for local AI, machine learning, CUDA development and technical visualization.\u003c\/p\u003e\u003ch3\u003eAI \u0026amp; ML use\u003c\/h3\u003e\u003cul\u003e\n\u003cli\u003eLocal LLM inference and AI assistants\u003c\/li\u003e\n\u003cli\u003eFine-tuning selected models\u003c\/li\u003e\n\u003cli\u003eComputer vision and deep learning\u003c\/li\u003e\n\u003cli\u003eDiffusion models and generative AI\u003c\/li\u003e\n\u003cli\u003ePyTorch, TensorFlow and CUDA development\u003c\/li\u003e\n\u003cli\u003eCAD\/CAE, rendering and visualization\u003c\/li\u003e\n\u003c\/ul\u003e\u003ch3\u003eWhy GPU memory matters\u003c\/h3\u003e\u003cp\u003e24 GB of VRAM is a practical capacity for many local LLM, computer-vision and generative-AI workflows. Whether a specific model fits also depends on precision, quantization, context length and batch size.\u003c\/p\u003e\u003cp\u003e\u003cstrong\u003eBest suited to:\u003c\/strong\u003e professional local AI development and combined AI\/engineering workflows.\u003c\/p\u003e","brand":"NVIDIA","offers":[{"title":"Default Title","offer_id":46838346514534,"sku":"900-5G195-2200-000","price":3310.0,"currency_code":"EUR","in_stock":false}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0680\/6572\/2470\/files\/nvidia-rtx-pro-4000-sff-blackwell-product.png?v=1788276910"},{"product_id":"lenovo-thinkstation-p8-ryzen-threadripper-pro-9975wx-128-gb-ecc-1-tb-nvme","title":"Lenovo ThinkStation P8 – Ryzen Threadripper PRO 9975WX – 128 GB ECC – 1 TB NVMe","description":"\u003cp\u003eProfessional high-end desktop workstation for CPU-intensive engineering, simulation, data and HPC workloads, and as an expandable foundation for AI systems.\u003c\/p\u003e\u003ch3\u003eAI \u0026amp; ML role\u003c\/h3\u003e\u003cul\u003e\n\u003cli\u003eDataset preparation and preprocessing\u003c\/li\u003e\n\u003cli\u003eCPU-based machine-learning and inference workloads\u003c\/li\u003e\n\u003cli\u003eLarge datasets and memory-intensive development\u003c\/li\u003e\n\u003cli\u003eSimulation, CAE\/CFD and parallel engineering workloads\u003c\/li\u003e\n\u003c\/ul\u003e\u003cp\u003e\u003cstrong\u003eImportant for GPU AI:\u003c\/strong\u003e the listed configuration does not show a dedicated NVIDIA AI GPU in the product title. CUDA-based training or fast local LLM inference therefore requires a compatible GPU configuration.\u003c\/p\u003e","brand":"Lenovo","offers":[{"title":"Default Title","offer_id":46844501098598,"sku":"30HH009UGE","price":12430.0,"currency_code":"EUR","in_stock":false}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0680\/6572\/2470\/files\/lenovo-thinkstation-p8-30hh009uge-front.jpg?v=1788279612"},{"product_id":"lenovo-thinkstation-p5-gen-2-xeon-638-rtx-pro-4500-32-gb-64-gb-ecc-1-tb-nvme","title":"Lenovo ThinkStation P5 Gen 2 – Xeon 638 – RTX PRO 4500 32 GB – 64 GB ECC – 1 TB NVMe","description":"\u003cp\u003eProfessional desktop workstation for local AI, machine learning, CUDA and engineering. The combination of a capable CPU and NVIDIA GPU suits developers who want AI and technical applications on one system.\u003c\/p\u003e\u003ch3\u003eAI \u0026amp; ML use\u003c\/h3\u003e\u003cul\u003e\n\u003cli\u003eLocal LLM inference and AI assistants\u003c\/li\u003e\n\u003cli\u003ePyTorch, TensorFlow and CUDA development\u003c\/li\u003e\n\u003cli\u003eComputer vision and generative AI\u003c\/li\u003e\n\u003cli\u003eRAG, data science and model testing\u003c\/li\u003e\n\u003cli\u003eCAD\/CAE, rendering and visualization\u003c\/li\u003e\n\u003c\/ul\u003e\u003cp\u003e\u003cstrong\u003eGPU memory:\u003c\/strong\u003e 32 GB of GPU memory provides substantial headroom for local AI models, larger datasets and selected fine-tuning tasks. Whether a specific model fits completely depends on precision, quantization, context length and batch size.\u003c\/p\u003e","brand":"Lenovo","offers":[{"title":"Default Title","offer_id":46844998352998,"sku":"30KB0028GE","price":9530.0,"currency_code":"EUR","in_stock":false}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0680\/6572\/2470\/files\/lenovo-thinkstation-p5-gen2-30kb0028ge-front.jpg?v=1788280058"},{"product_id":"lenovo-thinkstation-p4-ryzen-9-pro-9965x3d-rtx-pro-4000-24-gb-64-gb-1-tb-nvme","title":"Lenovo ThinkStation P4 – Ryzen 9 PRO 9965X3D – RTX PRO 4000 24 GB – 64 GB – 1 TB NVMe","description":"\u003cp\u003eProfessional desktop workstation for local AI, machine learning, CUDA and engineering. The combination of a capable CPU and NVIDIA GPU suits developers who want AI and technical applications on one system.\u003c\/p\u003e\u003ch3\u003eAI \u0026amp; ML use\u003c\/h3\u003e\u003cul\u003e\n\u003cli\u003eLocal LLM inference and AI assistants\u003c\/li\u003e\n\u003cli\u003ePyTorch, TensorFlow and CUDA development\u003c\/li\u003e\n\u003cli\u003eComputer vision and generative AI\u003c\/li\u003e\n\u003cli\u003eRAG, data science and model testing\u003c\/li\u003e\n\u003cli\u003eCAD\/CAE, rendering and visualization\u003c\/li\u003e\n\u003c\/ul\u003e\u003cp\u003e\u003cstrong\u003eGPU memory:\u003c\/strong\u003e 24 GPU memory is suitable for local inference, computer vision and professional development workflows. Whether a specific model fits completely depends on precision, quantization, context length and batch size.\u003c\/p\u003e","brand":"Lenovo","offers":[{"title":"Default Title","offer_id":46845997875302,"sku":"30KN003CGE","price":6400.0,"currency_code":"EUR","in_stock":false}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0680\/6572\/2470\/files\/lenovo-thinkstation-p4-30kn003cge-front.jpg?v=1788280786"},{"product_id":"amd-epyc-9965-192-core-server-processor","title":"AMD EPYC 9965 – 192 Cores \/ 384 Threads – 2.25–3.7 GHz – 384 MB L3 – Socket SP5","description":"\u003cp\u003eProfessional AMD EPYC server processor for AI\/HPC, data, virtualization and engineering systems. With 192 cores \/ 384 threads it handles much of the data- and memory-intensive work around accelerators in an AI system.\u003c\/p\u003e\u003ch3\u003eAI \u0026amp; machine learning use\u003c\/h3\u003e\u003cul\u003e\n\u003cli\u003eDataset preparation, preprocessing and feature engineering\u003c\/li\u003e\n\u003cli\u003eKeeping GPU accelerators supplied with data\u003c\/li\u003e\n\u003cli\u003eCPU inference and traditional machine-learning workloads\u003c\/li\u003e\n\u003cli\u003eContainers, virtualization and multi-user AI servers\u003c\/li\u003e\n\u003cli\u003eScientific computing, CFD, CAE and other HPC workloads\u003c\/li\u003e\n\u003c\/ul\u003e\u003ch3\u003eWhy this CPU profile?\u003c\/h3\u003e\u003cp\u003eThe high core count targets highly parallel data processing, virtualization and HPC throughput.\u003c\/p\u003e\u003cp\u003e\u003cstrong\u003eBest suited to:\u003c\/strong\u003e professional AI servers, HPC nodes and CPU\/GPU systems where memory bandwidth, parallelism and I\/O matter alongside accelerator performance.\u003c\/p\u003e","brand":"AMD","offers":[{"title":"Default Title","offer_id":46846767726694,"sku":"100-000000976","price":9490.0,"currency_code":"EUR","in_stock":false}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0680\/6572\/2470\/files\/amd-epyc-current-official-hero_20b41c87-ba6e-4ee9-94fd-f6ef16bd4388.jpg?v=1789206714"},{"product_id":"amd-epyc-9845-160-cores-320-threads-2-1-3-7-ghz-320-mb-l3-socket-sp5","title":"AMD EPYC 9845 – 160 Cores \/ 320 Threads – 2.1–3.7 GHz – 320 MB L3 – Socket SP5","description":"\u003cp\u003eProfessional AMD EPYC server processor for AI\/HPC, data, virtualization and engineering systems. With 160 cores \/ 320 threads it handles much of the data- and memory-intensive work around accelerators in an AI system.\u003c\/p\u003e\u003ch3\u003eAI \u0026amp; machine learning use\u003c\/h3\u003e\u003cul\u003e\n\u003cli\u003eDataset preparation, preprocessing and feature engineering\u003c\/li\u003e\n\u003cli\u003eKeeping GPU accelerators supplied with data\u003c\/li\u003e\n\u003cli\u003eCPU inference and traditional machine-learning workloads\u003c\/li\u003e\n\u003cli\u003eContainers, virtualization and multi-user AI servers\u003c\/li\u003e\n\u003cli\u003eScientific computing, CFD, CAE and other HPC workloads\u003c\/li\u003e\n\u003c\/ul\u003e\u003ch3\u003eWhy this CPU profile?\u003c\/h3\u003e\u003cp\u003eThe high core count targets highly parallel data processing, virtualization and HPC throughput.\u003c\/p\u003e\u003cp\u003e\u003cstrong\u003eBest suited to:\u003c\/strong\u003e professional AI servers, HPC nodes and CPU\/GPU systems where memory bandwidth, parallelism and I\/O matter alongside accelerator performance.\u003c\/p\u003e","brand":"AMD","offers":[{"title":"Default Title","offer_id":46847288901734,"sku":"100-000001458","price":8710.0,"currency_code":"EUR","in_stock":false}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0680\/6572\/2470\/files\/amd-epyc-current-official-hero_e853ef4e-b4f4-4142-a439-3412e078a9fa.jpg?v=1789206707"},{"product_id":"amd-epyc-9684x-96-cores-192-threads-2-55-3-7-ghz-1152-mb-l3-socket-sp5","title":"AMD EPYC 9684X – 96 Cores \/ 192 Threads – 2.55–3.7 GHz – 1152 MB L3 – Socket SP5","description":"\u003cp\u003eProfessional AMD EPYC server processor for AI\/HPC, data, virtualization and engineering systems. With 96 cores \/ 192 threads it handles much of the data- and memory-intensive work around accelerators in an AI system.\u003c\/p\u003e\u003ch3\u003eAI \u0026amp; machine learning use\u003c\/h3\u003e\u003cul\u003e\n\u003cli\u003eDataset preparation, preprocessing and feature engineering\u003c\/li\u003e\n\u003cli\u003eKeeping GPU accelerators supplied with data\u003c\/li\u003e\n\u003cli\u003eCPU inference and traditional machine-learning workloads\u003c\/li\u003e\n\u003cli\u003eContainers, virtualization and multi-user AI servers\u003c\/li\u003e\n\u003cli\u003eScientific computing, CFD, CAE and other HPC workloads\u003c\/li\u003e\n\u003c\/ul\u003e\u003ch3\u003eWhy this CPU profile?\u003c\/h3\u003e\u003cp\u003eThe very large L3 cache is particularly relevant to cache-sensitive simulation, scientific codes and data-intensive analytics.\u003c\/p\u003e\u003cp\u003e\u003cstrong\u003eBest suited to:\u003c\/strong\u003e professional AI servers, HPC nodes and CPU\/GPU systems where memory bandwidth, parallelism and I\/O matter alongside accelerator performance.\u003c\/p\u003e","brand":"AMD","offers":[{"title":"Default Title","offer_id":46853349900390,"sku":"100-000001254","price":8650.0,"currency_code":"EUR","in_stock":false}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0680\/6572\/2470\/files\/amd-epyc-current-official-hero_e5343235-bf99-4843-b572-3031953220c9.jpg?v=1789206669"},{"product_id":"amd-epyc-9755-128-cores-256-threads-2-7-4-1-ghz-512-mb-l3-socket-sp5","title":"AMD EPYC 9755 – 128 Cores \/ 256 Threads – 2.7–4.1 GHz – 512 MB L3 – Socket SP5","description":"\u003cp\u003eProfessional AMD EPYC server processor for AI\/HPC, data, virtualization and engineering systems. With 128 cores \/ 256 threads it handles much of the data- and memory-intensive work around accelerators in an AI system.\u003c\/p\u003e\u003ch3\u003eAI \u0026amp; machine learning use\u003c\/h3\u003e\u003cul\u003e\n\u003cli\u003eDataset preparation, preprocessing and feature engineering\u003c\/li\u003e\n\u003cli\u003eKeeping GPU accelerators supplied with data\u003c\/li\u003e\n\u003cli\u003eCPU inference and traditional machine-learning workloads\u003c\/li\u003e\n\u003cli\u003eContainers, virtualization and multi-user AI servers\u003c\/li\u003e\n\u003cli\u003eScientific computing, CFD, CAE and other HPC workloads\u003c\/li\u003e\n\u003c\/ul\u003e\u003ch3\u003eWhy this CPU profile?\u003c\/h3\u003e\u003cp\u003eThe high core count targets highly parallel data processing, virtualization and HPC throughput.\u003c\/p\u003e\u003cp\u003e\u003cstrong\u003eBest suited to:\u003c\/strong\u003e professional AI servers, HPC nodes and CPU\/GPU systems where memory bandwidth, parallelism and I\/O matter alongside accelerator performance.\u003c\/p\u003e","brand":"AMD","offers":[{"title":"Default Title","offer_id":46853825298534,"sku":"100-000001443","price":8350.0,"currency_code":"EUR","in_stock":false}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0680\/6572\/2470\/files\/amd-epyc-current-official-hero_98f9bc8c-7dab-41bf-8cc1-a8a377c79113.jpg?v=1789206695"},{"product_id":"amd-epyc-9825-144-cores-288-threads-2-2-3-7-ghz-384-mb-l3-socket-sp5","title":"AMD EPYC 9825 – 144 Cores \/ 288 Threads – 2.2–3.7 GHz – 384 MB L3 – Socket SP5","description":"\u003cp\u003eProfessional AMD EPYC server processor for AI\/HPC, data, virtualization and engineering systems. With 144 cores \/ 288 threads it handles much of the data- and memory-intensive work around accelerators in an AI system.\u003c\/p\u003e\u003ch3\u003eAI \u0026amp; machine learning use\u003c\/h3\u003e\u003cul\u003e\n\u003cli\u003eDataset preparation, preprocessing and feature engineering\u003c\/li\u003e\n\u003cli\u003eKeeping GPU accelerators supplied with data\u003c\/li\u003e\n\u003cli\u003eCPU inference and traditional machine-learning workloads\u003c\/li\u003e\n\u003cli\u003eContainers, virtualization and multi-user AI servers\u003c\/li\u003e\n\u003cli\u003eScientific computing, CFD, CAE and other HPC workloads\u003c\/li\u003e\n\u003c\/ul\u003e\u003ch3\u003eWhy this CPU profile?\u003c\/h3\u003e\u003cp\u003eThe high core count targets highly parallel data processing, virtualization and HPC throughput.\u003c\/p\u003e\u003cp\u003e\u003cstrong\u003eBest suited to:\u003c\/strong\u003e professional AI servers, HPC nodes and CPU\/GPU systems where memory bandwidth, parallelism and I\/O matter alongside accelerator performance.\u003c\/p\u003e","brand":"AMD","offers":[{"title":"Default Title","offer_id":46896624795750,"sku":"100-000000837","price":7890.0,"currency_code":"EUR","in_stock":false}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0680\/6572\/2470\/files\/amd-epyc-current-official-hero_04b77074-6095-451e-862f-83896615c3cb.jpg?v=1789206702"},{"product_id":"amd-epyc-9745-128-cores-256-threads-2-4-3-7-ghz-256-mb-l3-socket-sp5","title":"AMD EPYC 9745 – 128 Cores \/ 256 Threads – 2.4–3.7 GHz – 256 MB L3 – Socket SP5","description":"\u003cp\u003eProfessional AMD EPYC server processor for AI\/HPC, data, virtualization and engineering systems. With 128 cores \/ 256 threads it handles much of the data- and memory-intensive work around accelerators in an AI system.\u003c\/p\u003e\u003ch3\u003eAI \u0026amp; machine learning use\u003c\/h3\u003e\u003cul\u003e\n\u003cli\u003eDataset preparation, preprocessing and feature engineering\u003c\/li\u003e\n\u003cli\u003eKeeping GPU accelerators supplied with data\u003c\/li\u003e\n\u003cli\u003eCPU inference and traditional machine-learning workloads\u003c\/li\u003e\n\u003cli\u003eContainers, virtualization and multi-user AI servers\u003c\/li\u003e\n\u003cli\u003eScientific computing, CFD, CAE and other HPC workloads\u003c\/li\u003e\n\u003c\/ul\u003e\u003ch3\u003eWhy this CPU profile?\u003c\/h3\u003e\u003cp\u003eThe high core count targets highly parallel data processing, virtualization and HPC throughput.\u003c\/p\u003e\u003cp\u003e\u003cstrong\u003eBest suited to:\u003c\/strong\u003e professional AI servers, HPC nodes and CPU\/GPU systems where memory bandwidth, parallelism and I\/O matter alongside accelerator performance.\u003c\/p\u003e","brand":"AMD","offers":[{"title":"Default Title","offer_id":46896637870182,"sku":"100-000001460","price":7820.0,"currency_code":"EUR","in_stock":false}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0680\/6572\/2470\/files\/amd-epyc-current-official-hero_90044fab-ac0b-4faf-bc54-ff09ade215ba.jpg?v=1789206680"},{"product_id":"amd-epyc-9655-96-cores-192-threads-2-6-4-5-ghz-384-mb-l3-socket-sp5","title":"AMD EPYC 9655 – 96 Cores \/ 192 Threads – 2.6–4.5 GHz – 384 MB L3 – Socket SP5","description":"\u003cp\u003eProfessional AMD EPYC server processor for AI\/HPC, data, virtualization and engineering systems. With 96 cores \/ 192 threads it handles much of the data- and memory-intensive work around accelerators in an AI system.\u003c\/p\u003e\u003ch3\u003eAI \u0026amp; machine learning use\u003c\/h3\u003e\u003cul\u003e\n\u003cli\u003eDataset preparation, preprocessing and feature engineering\u003c\/li\u003e\n\u003cli\u003eKeeping GPU accelerators supplied with data\u003c\/li\u003e\n\u003cli\u003eCPU inference and traditional machine-learning workloads\u003c\/li\u003e\n\u003cli\u003eContainers, virtualization and multi-user AI servers\u003c\/li\u003e\n\u003cli\u003eScientific computing, CFD, CAE and other HPC workloads\u003c\/li\u003e\n\u003c\/ul\u003e\u003ch3\u003eWhy this CPU profile?\u003c\/h3\u003e\u003cp\u003eThe balance of core count and frequency suits mixed server, engineering and data workloads.\u003c\/p\u003e\u003cp\u003e\u003cstrong\u003eBest suited to:\u003c\/strong\u003e professional AI servers, HPC nodes and CPU\/GPU systems where memory bandwidth, parallelism and I\/O matter alongside accelerator performance.\u003c\/p\u003e","brand":"AMD","offers":[{"title":"Default Title","offer_id":46896766582886,"sku":"100-000000674","price":6880.0,"currency_code":"EUR","in_stock":false}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0680\/6572\/2470\/files\/amd-epyc-current-official-hero_b947165b-8f81-4406-a280-fa4dc9654ebe.jpg?v=1789206664"},{"product_id":"amd-epyc-9555p-64-cores-128-threads-3-2-4-4-ghz-256-mb-l3-socket-sp5","title":"AMD EPYC 9555P – 64 Cores \/ 128 Threads – 3.2–4.4 GHz – 256 MB L3 – Socket SP5","description":"\u003cp\u003eProfessional AMD EPYC server processor for AI\/HPC, data, virtualization and engineering systems. With 64 cores \/ 128 threads it handles much of the data- and memory-intensive work around accelerators in an AI system.\u003c\/p\u003e\u003ch3\u003eAI \u0026amp; machine learning use\u003c\/h3\u003e\u003cul\u003e\n\u003cli\u003eDataset preparation, preprocessing and feature engineering\u003c\/li\u003e\n\u003cli\u003eKeeping GPU accelerators supplied with data\u003c\/li\u003e\n\u003cli\u003eCPU inference and traditional machine-learning workloads\u003c\/li\u003e\n\u003cli\u003eContainers, virtualization and multi-user AI servers\u003c\/li\u003e\n\u003cli\u003eScientific computing, CFD, CAE and other HPC workloads\u003c\/li\u003e\n\u003c\/ul\u003e\u003ch3\u003eWhy this CPU profile?\u003c\/h3\u003e\u003cp\u003eThe balance of core count and frequency suits mixed server, engineering and data workloads.\u003c\/p\u003e\u003cp\u003e\u003cstrong\u003eBest suited to:\u003c\/strong\u003e professional AI servers, HPC nodes and CPU\/GPU systems where memory bandwidth, parallelism and I\/O matter alongside accelerator performance.\u003c\/p\u003e","brand":"AMD","offers":[{"title":"Default","offer_id":46896869146726,"sku":"100-000001523","price":5190.0,"currency_code":"EUR","in_stock":false}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0680\/6572\/2470\/files\/amd-epyc-current-official-hero_aad1bc2c-66be-44fa-afba-c8b95b01ed82.jpg?v=1789206642"},{"product_id":"amd-epyc-9754-128-cores-256-threads-2-25-3-1-ghz-256-mb-l3-socket-sp5","title":"AMD EPYC 9754 – 128 Cores \/ 256 Threads – 2.25–3.1 GHz – 256 MB L3 – Socket SP5","description":"\u003cp\u003eProfessional AMD EPYC server processor for AI\/HPC, data, virtualization and engineering systems. With 128 cores \/ 256 threads it handles much of the data- and memory-intensive work around accelerators in an AI system.\u003c\/p\u003e\u003ch3\u003eAI \u0026amp; machine learning use\u003c\/h3\u003e\u003cul\u003e\n\u003cli\u003eDataset preparation, preprocessing and feature engineering\u003c\/li\u003e\n\u003cli\u003eKeeping GPU accelerators supplied with data\u003c\/li\u003e\n\u003cli\u003eCPU inference and traditional machine-learning workloads\u003c\/li\u003e\n\u003cli\u003eContainers, virtualization and multi-user AI servers\u003c\/li\u003e\n\u003cli\u003eScientific computing, CFD, CAE and other HPC workloads\u003c\/li\u003e\n\u003c\/ul\u003e\u003ch3\u003eWhy this CPU profile?\u003c\/h3\u003e\u003cp\u003eThe high core count targets highly parallel data processing, virtualization and HPC throughput.\u003c\/p\u003e\u003cp\u003e\u003cstrong\u003eBest suited to:\u003c\/strong\u003e professional AI servers, HPC nodes and CPU\/GPU systems where memory bandwidth, parallelism and I\/O matter alongside accelerator performance.\u003c\/p\u003e","brand":"AMD","offers":[{"title":"Default","offer_id":46896878878822,"sku":"100-000001234","price":3890.0,"currency_code":"EUR","in_stock":false}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0680\/6572\/2470\/files\/amd-epyc-current-official-hero_d49eb361-546c-4fe9-b5c4-cd25e500d7fb.jpg?v=1789206687"},{"product_id":"amd-epyc-9654-96-cores-192-threads-2-4-3-7-ghz-384-mb-l3-socket-sp5","title":"AMD EPYC 9654 – 96 Cores \/ 192 Threads – 2.4–3.7 GHz – 384 MB L3 – Socket SP5","description":"\u003cp\u003eProfessional AMD EPYC server processor for AI\/HPC, data, virtualization and engineering systems. With 96 cores \/ 192 threads it handles much of the data- and memory-intensive work around accelerators in an AI system.\u003c\/p\u003e\u003ch3\u003eAI \u0026amp; machine learning use\u003c\/h3\u003e\u003cul\u003e\n\u003cli\u003eDataset preparation, preprocessing and feature engineering\u003c\/li\u003e\n\u003cli\u003eKeeping GPU accelerators supplied with data\u003c\/li\u003e\n\u003cli\u003eCPU inference and traditional machine-learning workloads\u003c\/li\u003e\n\u003cli\u003eContainers, virtualization and multi-user AI servers\u003c\/li\u003e\n\u003cli\u003eScientific computing, CFD, CAE and other HPC workloads\u003c\/li\u003e\n\u003c\/ul\u003e\u003ch3\u003eWhy this CPU profile?\u003c\/h3\u003e\u003cp\u003eThe balance of core count and frequency suits mixed server, engineering and data workloads.\u003c\/p\u003e\u003cp\u003e\u003cstrong\u003eBest suited to:\u003c\/strong\u003e professional AI servers, HPC nodes and CPU\/GPU systems where memory bandwidth, parallelism and I\/O matter alongside accelerator performance.\u003c\/p\u003e","brand":"AMD","offers":[{"title":"Default Title","offer_id":46896887398502,"sku":"100-000000789","price":3460.0,"currency_code":"EUR","in_stock":false}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0680\/6572\/2470\/files\/amd-epyc-current-official-hero_22cc6bd9-c17b-4fc0-9846-ea56fa34307c.jpg?v=1789206651"},{"product_id":"amd-epyc-9355-32-cores-64-threads-3-55-4-4-ghz-256-mb-l3-socket-sp5","title":"AMD EPYC 9355 – 32 Cores \/ 64 Threads – 3.55–4.4 GHz – 256 MB L3 – Socket SP5","description":"\u003cp\u003eProfessional AMD EPYC server processor for AI\/HPC, data, virtualization and engineering systems. With 32 cores \/ 64 threads it handles much of the data- and memory-intensive work around accelerators in an AI system.\u003c\/p\u003e\u003ch3\u003eAI \u0026amp; machine learning use\u003c\/h3\u003e\u003cul\u003e\n\u003cli\u003eDataset preparation, preprocessing and feature engineering\u003c\/li\u003e\n\u003cli\u003eKeeping GPU accelerators supplied with data\u003c\/li\u003e\n\u003cli\u003eCPU inference and traditional machine-learning workloads\u003c\/li\u003e\n\u003cli\u003eContainers, virtualization and multi-user AI servers\u003c\/li\u003e\n\u003cli\u003eScientific computing, CFD, CAE and other HPC workloads\u003c\/li\u003e\n\u003c\/ul\u003e\u003ch3\u003eWhy this CPU profile?\u003c\/h3\u003e\u003cp\u003eThe balance of core count and frequency suits mixed server, engineering and data workloads.\u003c\/p\u003e\u003cp\u003e\u003cstrong\u003eBest suited to:\u003c\/strong\u003e professional AI servers, HPC nodes and CPU\/GPU systems where memory bandwidth, parallelism and I\/O matter alongside accelerator performance.\u003c\/p\u003e","brand":"AMD","offers":[{"title":"Default Title","offer_id":46896898965606,"sku":"100-000001148","price":3430.0,"currency_code":"EUR","in_stock":false}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0680\/6572\/2470\/files\/amd-epyc-current-official-hero_d5bc2451-121b-4ccb-803f-4edacaf7b484.jpg?v=1789206600"},{"product_id":"amd-epyc-9654p-96-cores-192-threads-2-4-3-7-ghz-384-mb-l3-socket-sp5","title":"AMD EPYC 9654P – 96 Cores \/ 192 Threads – 2.4–3.7 GHz – 384 MB L3 – Socket SP5","description":"\u003cp\u003eProfessional AMD EPYC server processor for AI\/HPC, data, virtualization and engineering systems. With 96 cores \/ 192 threads it handles much of the data- and memory-intensive work around accelerators in an AI system.\u003c\/p\u003e\u003ch3\u003eAI \u0026amp; machine learning use\u003c\/h3\u003e\u003cul\u003e\n\u003cli\u003eDataset preparation, preprocessing and feature engineering\u003c\/li\u003e\n\u003cli\u003eKeeping GPU accelerators supplied with data\u003c\/li\u003e\n\u003cli\u003eCPU inference and traditional machine-learning workloads\u003c\/li\u003e\n\u003cli\u003eContainers, virtualization and multi-user AI servers\u003c\/li\u003e\n\u003cli\u003eScientific computing, CFD, CAE and other HPC workloads\u003c\/li\u003e\n\u003c\/ul\u003e\u003ch3\u003eWhy this CPU profile?\u003c\/h3\u003e\u003cp\u003eThe balance of core count and frequency suits mixed server, engineering and data workloads.\u003c\/p\u003e\u003cp\u003e\u003cstrong\u003eBest suited to:\u003c\/strong\u003e professional AI servers, HPC nodes and CPU\/GPU systems where memory bandwidth, parallelism and I\/O matter alongside accelerator performance.\u003c\/p\u003e","brand":"AMD","offers":[{"title":"Default Title","offer_id":46896911417446,"sku":"100-000000803","price":3350.0,"currency_code":"EUR","in_stock":false}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0680\/6572\/2470\/files\/amd-epyc-current-official-hero_c8942eef-9767-46d7-b971-53b90120726b.jpg?v=1789206657"},{"product_id":"amd-epyc-9554-64-cores-128-threads-3-1-3-75-ghz-256-mb-l3-socket-sp5","title":"AMD EPYC 9554 – 64 Cores \/ 128 Threads – 3.1–3.75 GHz – 256 MB L3 – Socket SP5","description":"\u003cp\u003eProfessional AMD EPYC server processor for AI\/HPC, data, virtualization and engineering systems. With 64 cores \/ 128 threads it handles much of the data- and memory-intensive work around accelerators in an AI system.\u003c\/p\u003e\u003ch3\u003eAI \u0026amp; machine learning use\u003c\/h3\u003e\u003cul\u003e\n\u003cli\u003eDataset preparation, preprocessing and feature engineering\u003c\/li\u003e\n\u003cli\u003eKeeping GPU accelerators supplied with data\u003c\/li\u003e\n\u003cli\u003eCPU inference and traditional machine-learning workloads\u003c\/li\u003e\n\u003cli\u003eContainers, virtualization and multi-user AI servers\u003c\/li\u003e\n\u003cli\u003eScientific computing, CFD, CAE and other HPC workloads\u003c\/li\u003e\n\u003c\/ul\u003e\u003ch3\u003eWhy this CPU profile?\u003c\/h3\u003e\u003cp\u003eThe balance of core count and frequency suits mixed server, engineering and data workloads.\u003c\/p\u003e\u003cp\u003e\u003cstrong\u003eBest suited to:\u003c\/strong\u003e professional AI servers, HPC nodes and CPU\/GPU systems where memory bandwidth, parallelism and I\/O matter alongside accelerator performance.\u003c\/p\u003e","brand":"AMD","offers":[{"title":"Default Title","offer_id":46896925966438,"sku":"100-000000790","price":2920.0,"currency_code":"EUR","in_stock":false}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0680\/6572\/2470\/files\/amd-epyc-current-official-hero_a03c6c17-d440-4d87-bc4d-c9111dd43a0c.jpg?v=1789206626"},{"product_id":"amd-epyc-9474f-48-cores-96-threads-3-6-4-1-ghz-256-mb-l3-socket-sp5","title":"AMD EPYC 9474F – 48 Cores \/ 96 Threads – 3.6–4.1 GHz – 256 MB L3 – Socket SP5","description":"\u003cp\u003eProfessional AMD EPYC server processor for AI\/HPC, data, virtualization and engineering systems. With 48 cores \/ 96 threads it handles much of the data- and memory-intensive work around accelerators in an AI system.\u003c\/p\u003e\u003ch3\u003eAI \u0026amp; machine learning use\u003c\/h3\u003e\u003cul\u003e\n\u003cli\u003eDataset preparation, preprocessing and feature engineering\u003c\/li\u003e\n\u003cli\u003eKeeping GPU accelerators supplied with data\u003c\/li\u003e\n\u003cli\u003eCPU inference and traditional machine-learning workloads\u003c\/li\u003e\n\u003cli\u003eContainers, virtualization and multi-user AI servers\u003c\/li\u003e\n\u003cli\u003eScientific computing, CFD, CAE and other HPC workloads\u003c\/li\u003e\n\u003c\/ul\u003e\u003ch3\u003eWhy this CPU profile?\u003c\/h3\u003e\u003cp\u003eThe high-frequency profile is useful when strong per-core performance matters more than maximum core count.\u003c\/p\u003e\u003cp\u003e\u003cstrong\u003eBest suited to:\u003c\/strong\u003e professional AI servers, HPC nodes and CPU\/GPU systems where memory bandwidth, parallelism and I\/O matter alongside accelerator performance.\u003c\/p\u003e","brand":"AMD","offers":[{"title":"Default Title","offer_id":46896941858918,"sku":"100-000000788","price":2630.0,"currency_code":"EUR","in_stock":false}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0680\/6572\/2470\/files\/amd-epyc-current-official-hero_8ed29bab-bee4-4068-8bc1-01c7a092b199.jpg?v=1789206617"},{"product_id":"amd-epyc-9554p-64-cores-128-threads-3-1-3-75-ghz-256-mb-l3-socket-sp5","title":"AMD EPYC 9554P – 64 Cores \/ 128 Threads – 3.1–3.75 GHz – 256 MB L3 – Socket SP5","description":"\u003cp\u003eProfessional AMD EPYC server processor for AI\/HPC, data, virtualization and engineering systems. With 64 cores \/ 128 threads it handles much of the data- and memory-intensive work around accelerators in an AI system.\u003c\/p\u003e\u003ch3\u003eAI \u0026amp; machine learning use\u003c\/h3\u003e\u003cul\u003e\n\u003cli\u003eDataset preparation, preprocessing and feature engineering\u003c\/li\u003e\n\u003cli\u003eKeeping GPU accelerators supplied with data\u003c\/li\u003e\n\u003cli\u003eCPU inference and traditional machine-learning workloads\u003c\/li\u003e\n\u003cli\u003eContainers, virtualization and multi-user AI servers\u003c\/li\u003e\n\u003cli\u003eScientific computing, CFD, CAE and other HPC workloads\u003c\/li\u003e\n\u003c\/ul\u003e\u003ch3\u003eWhy this CPU profile?\u003c\/h3\u003e\u003cp\u003eThe balance of core count and frequency suits mixed server, engineering and data workloads.\u003c\/p\u003e\u003cp\u003e\u003cstrong\u003eBest suited to:\u003c\/strong\u003e professional AI servers, HPC nodes and CPU\/GPU systems where memory bandwidth, parallelism and I\/O matter alongside accelerator performance.\u003c\/p\u003e","brand":"AMD","offers":[{"title":"Default Title","offer_id":46896948805734,"sku":"100-000000804","price":2630.0,"currency_code":"EUR","in_stock":false}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0680\/6572\/2470\/files\/amd-epyc-current-official-hero_898d9849-a451-4b4c-b217-51a47432ff4d.jpg?v=1789206633"},{"product_id":"amd-epyc-9374f-32-cores-64-threads-3-85-4-3-ghz-256-mb-l3-socket-sp5","title":"AMD EPYC 9374F – 32 Cores \/ 64 Threads – 3.85–4.3 GHz – 256 MB L3 – Socket SP5","description":"\u003cp\u003eProfessional AMD EPYC server processor for AI\/HPC, data, virtualization and engineering systems. With 32 cores \/ 64 threads it handles much of the data- and memory-intensive work around accelerators in an AI system.\u003c\/p\u003e\u003ch3\u003eAI \u0026amp; machine learning use\u003c\/h3\u003e\u003cul\u003e\n\u003cli\u003eDataset preparation, preprocessing and feature engineering\u003c\/li\u003e\n\u003cli\u003eKeeping GPU accelerators supplied with data\u003c\/li\u003e\n\u003cli\u003eCPU inference and traditional machine-learning workloads\u003c\/li\u003e\n\u003cli\u003eContainers, virtualization and multi-user AI servers\u003c\/li\u003e\n\u003cli\u003eScientific computing, CFD, CAE and other HPC workloads\u003c\/li\u003e\n\u003c\/ul\u003e\u003ch3\u003eWhy this CPU profile?\u003c\/h3\u003e\u003cp\u003eThe high-frequency profile is useful when strong per-core performance matters more than maximum core count.\u003c\/p\u003e\u003cp\u003e\u003cstrong\u003eBest suited to:\u003c\/strong\u003e professional AI servers, HPC nodes and CPU\/GPU systems where memory bandwidth, parallelism and I\/O matter alongside accelerator performance.\u003c\/p\u003e","brand":"AMD","offers":[{"title":"Default Title","offer_id":46896957390950,"sku":"100-000000792","price":2520.0,"currency_code":"EUR","in_stock":false}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0680\/6572\/2470\/files\/amd-epyc-current-official-hero_04925c99-8b34-4b1c-8f98-eac93bfdf41c.jpg?v=1789206605"},{"product_id":"apple-macbook-pro-16-m5-max-48gb-2tb-space-black","title":"Apple MacBook Pro 16 M5 Max 48GB 2TB Space Black","description":"\u003cp\u003eHigh-performance mobile Apple system for local AI development, data science, media and engineering workflows. Its large unified-memory pool can be attractive for local models that fit within shared memory.\u003c\/p\u003e\u003ch3\u003eAI \u0026amp; ML use\u003c\/h3\u003e\u003cul\u003e\n\u003cli\u003eLocal LLM inference with Apple-optimized frameworks\u003c\/li\u003e\n\u003cli\u003eMLX, Metal and PyTorch MPS workflows\u003c\/li\u003e\n\u003cli\u003eData science, development and prototyping\u003c\/li\u003e\n\u003cli\u003eComputer vision and generative workflows with compatible software\u003c\/li\u003e\n\u003c\/ul\u003e\u003cp\u003e\u003cstrong\u003eImportant:\u003c\/strong\u003e macOS does not use NVIDIA CUDA. Check that your frameworks and models support Metal, MLX or MPS.\u003c\/p\u003e","brand":"Apple","offers":[{"title":"Default Title","offer_id":47053153927270,"sku":"MGEE4D\/A","price":5850.0,"currency_code":"EUR","in_stock":false}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0680\/6572\/2470\/files\/apple-macbook-pro-m5-max-mgee4da-01.jpg?v=1788544405"},{"product_id":"asus-rog-nuc-15-core-ultra-9-275hx-geforce-rtx-5080-32gb-2tb-ssd","title":"ASUS ROG NUC 15 – Core Ultra 9 275HX – GeForce RTX 5080 – 32GB – 2TB SSD","description":"\u003cp\u003eCompact NVIDIA RTX workstation for local AI development, CUDA, data science and technical applications.\u003c\/p\u003e\u003ch3\u003eAI \u0026amp; ML use\u003c\/h3\u003e\u003cul\u003e\n\u003cli\u003eLocal LLM inference and AI assistants\u003c\/li\u003e\n\u003cli\u003ePyTorch, TensorFlow and CUDA development\u003c\/li\u003e\n\u003cli\u003eComputer vision, image processing and generative AI\u003c\/li\u003e\n\u003cli\u003eRAG, prototyping and model testing\u003c\/li\u003e\n\u003c\/ul\u003e\u003cp\u003eThe compact format works well as a personal development system. For larger models, the actual GPU memory and the thermal\/power limits of the specific configuration are especially important.\u003c\/p\u003e","brand":"ASUS","offers":[{"title":"Default Title","offer_id":47053161398374,"sku":"90AS00I1-M00090","price":4430.0,"currency_code":"EUR","in_stock":false}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0680\/6572\/2470\/files\/asus-rog-nuc-15-rnuc15jnk9x28aa2-01.jpg?v=1788545052"},{"product_id":"lenovo-thinkpad-t16g-gen-3-core-ultra-9-275hx-rtx-5080-16-gb-64-gb-1-tb","title":"Lenovo ThinkPad T16g Gen 3 – Core Ultra 9 275HX – RTX 5080 16 GB – 64 GB – 1 TB","description":"\u003cp\u003eHigh-performance mobile workstation for AI, machine-learning and engineering development on the move. The NVIDIA GPU enables CUDA-accelerated workflows directly on the notebook.\u003c\/p\u003e\u003ch3\u003eTypical AI workloads\u003c\/h3\u003e\u003cul\u003e\n\u003cli\u003ePython, PyTorch and CUDA development\u003c\/li\u003e\n\u003cli\u003eLocal LLM inference and AI assistants\u003c\/li\u003e\n\u003cli\u003eComputer vision and image analysis\u003c\/li\u003e\n\u003cli\u003eRAG demos, data science and prototyping\u003c\/li\u003e\n\u003cli\u003eCAD\/CAE, rendering and technical visualization\u003c\/li\u003e\n\u003c\/ul\u003e\u003cp\u003e\u003cstrong\u003eGPU memory:\u003c\/strong\u003e 16 GB of GPU memory is well suited to development, computer vision and many small-to-medium local AI workloads. Model size and speed also depend on precision, quantization and the notebook's power profile.\u003c\/p\u003e\u003cp\u003e\u003cstrong\u003eBest suited to:\u003c\/strong\u003e developers and engineers who value mobility more than the maximum sustained performance and expandability of a tower workstation.\u003c\/p\u003e","brand":"Lenovo","offers":[{"title":"Default Title","offer_id":47162765344870,"sku":"21V50006GE","price":5760.0,"currency_code":"EUR","in_stock":false}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0680\/6572\/2470\/files\/lenovo-thinkpad-t16g-gen3-21v50006ge-01.jpg?v=1789206137"},{"product_id":"lenovo-thinkpad-p16-gen-3-core-ultra-9-275hx-rtx-pro-2000-8-gb-64-gb-1-tb","title":"Lenovo ThinkPad P16 Gen 3 – Core Ultra 9 275HX – RTX PRO 2000 8 GB – 64 GB – 1 TB","description":"\u003cp\u003eHigh-performance mobile workstation for AI, machine-learning and engineering development on the move. The NVIDIA GPU enables CUDA-accelerated workflows directly on the notebook.\u003c\/p\u003e\u003ch3\u003eTypical AI workloads\u003c\/h3\u003e\u003cul\u003e\n\u003cli\u003ePython, PyTorch and CUDA development\u003c\/li\u003e\n\u003cli\u003eLocal LLM inference and AI assistants\u003c\/li\u003e\n\u003cli\u003eComputer vision and image analysis\u003c\/li\u003e\n\u003cli\u003eRAG demos, data science and prototyping\u003c\/li\u003e\n\u003cli\u003eCAD\/CAE, rendering and technical visualization\u003c\/li\u003e\n\u003c\/ul\u003e\u003cp\u003e\u003cstrong\u003eGPU memory:\u003c\/strong\u003e 8 GPU memory is better suited to development, computer vision, prototyping and smaller local models. Model size and speed also depend on precision, quantization and the notebook's power profile.\u003c\/p\u003e\u003cp\u003e\u003cstrong\u003eBest suited to:\u003c\/strong\u003e developers and engineers who value mobility more than the maximum sustained performance and expandability of a tower workstation.\u003c\/p\u003e","brand":"Lenovo","offers":[{"title":"Default Title","offer_id":47162844577894,"sku":"21RQ004QGE","price":5230.0,"currency_code":"EUR","in_stock":false}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0680\/6572\/2470\/files\/lenovo-thinkpad-p16-gen3-21rq004qge-01.jpg?v=1789206849"},{"product_id":"lenovo-thinkstation-p8-threadripper-pro-9965wx-128-gb-ecc-1-tb-nvme","title":"Lenovo ThinkStation P8 – Threadripper PRO 9965WX – 128 GB ECC – 1 TB NVMe","description":"\u003cp\u003eProfessional high-end desktop workstation for CPU-intensive engineering, simulation, data and HPC workloads, and as an expandable foundation for AI systems.\u003c\/p\u003e\u003ch3\u003eAI \u0026amp; ML role\u003c\/h3\u003e\u003cul\u003e\n\u003cli\u003eDataset preparation and preprocessing\u003c\/li\u003e\n\u003cli\u003eCPU-based machine-learning and inference workloads\u003c\/li\u003e\n\u003cli\u003eLarge datasets and memory-intensive development\u003c\/li\u003e\n\u003cli\u003eSimulation, CAE\/CFD and parallel engineering workloads\u003c\/li\u003e\n\u003c\/ul\u003e\u003cp\u003e\u003cstrong\u003eImportant for GPU AI:\u003c\/strong\u003e the listed configuration does not show a dedicated NVIDIA AI GPU in the product title. CUDA-based training or fast local LLM inference therefore requires a compatible GPU configuration.\u003c\/p\u003e","brand":"Lenovo","offers":[{"title":"Default Title","offer_id":47163460026470,"sku":"30HH00A3GE","price":10580.0,"currency_code":"EUR","in_stock":false}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0680\/6572\/2470\/files\/lenovo-thinkstation-p8-30hh00a3ge-01.jpg?v=1789207793"},{"product_id":"lenovo-thinkstation-p3-ultra-gen-2-core-ultra-9-285-rtx-pro-4000-sff-24-gb-64-gb-1-tb","title":"Lenovo ThinkStation P3 Ultra Gen 2 – Core Ultra 9 285 – RTX PRO 4000 SFF 24 GB – 64 GB – 1 TB","description":"\u003cp\u003eProfessional desktop workstation for local AI, machine learning, CUDA and engineering. The combination of a capable CPU and NVIDIA GPU suits developers who want AI and technical applications on one system.\u003c\/p\u003e\u003ch3\u003eAI \u0026amp; ML use\u003c\/h3\u003e\u003cul\u003e\n\u003cli\u003eLocal LLM inference and AI assistants\u003c\/li\u003e\n\u003cli\u003ePyTorch, TensorFlow and CUDA development\u003c\/li\u003e\n\u003cli\u003eComputer vision and generative AI\u003c\/li\u003e\n\u003cli\u003eRAG, data science and model testing\u003c\/li\u003e\n\u003cli\u003eCAD\/CAE, rendering and visualization\u003c\/li\u003e\n\u003c\/ul\u003e\u003cp\u003e\u003cstrong\u003eGPU memory:\u003c\/strong\u003e 24 GPU memory is suitable for local inference, computer vision and professional development workflows. Whether a specific model fits completely depends on precision, quantization, context length and batch size.\u003c\/p\u003e","brand":"Lenovo","offers":[{"title":"Default Title","offer_id":47163461075046,"sku":"30J50086GE","price":5410.0,"currency_code":"EUR","in_stock":false}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0680\/6572\/2470\/files\/lenovo-thinkstation-p3-ultra-gen2-30j50086ge-01.jpg?v=1789208105"},{"product_id":"lenovo-thinkstation-p3-gen-2-core-ultra-9-285k-rtx-pro-2000-blackwell-16-gb-64-gb-1-tb","title":"Lenovo ThinkStation P3 Gen 2 – Core Ultra 9 285K – RTX PRO 2000 Blackwell 16 GB – 64 GB – 1 TB","description":"\u003cp\u003eProfessional desktop workstation for local AI, machine learning, CUDA and engineering. The combination of a capable CPU and NVIDIA GPU suits developers who want AI and technical applications on one system.\u003c\/p\u003e\u003ch3\u003eAI \u0026amp; ML use\u003c\/h3\u003e\u003cul\u003e\n\u003cli\u003eLocal LLM inference and AI assistants\u003c\/li\u003e\n\u003cli\u003ePyTorch, TensorFlow and CUDA development\u003c\/li\u003e\n\u003cli\u003eComputer vision and generative AI\u003c\/li\u003e\n\u003cli\u003eRAG, data science and model testing\u003c\/li\u003e\n\u003cli\u003eCAD\/CAE, rendering and visualization\u003c\/li\u003e\n\u003c\/ul\u003e\u003cp\u003e\u003cstrong\u003eGPU memory:\u003c\/strong\u003e 16 GPU memory is suitable for local inference, computer vision and professional development workflows. Whether a specific model fits completely depends on precision, quantization, context length and batch size.\u003c\/p\u003e","brand":"Lenovo","offers":[{"title":"Default Title","offer_id":47163461730406,"sku":"30HT00AHGE","price":4190.0,"currency_code":"EUR","in_stock":false}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0680\/6572\/2470\/files\/lenovo-thinkstation-p3-gen2-30ht00ahge-01.jpg?v=1789208378"},{"product_id":"nvidia-rtx-pro-4500-blackwell-retail-32-gb-gddr7-ecc-pcie-5-0","title":"NVIDIA RTX PRO 4500 Blackwell Retail – 32 GB GDDR7 ECC – PCIe 5.0","description":"\u003cp\u003eProfessional NVIDIA RTX PRO workstation GPU with substantial GB of graphics memory for local AI, machine learning, CUDA development and technical visualization.\u003c\/p\u003e\u003ch3\u003eAI \u0026amp; ML use\u003c\/h3\u003e\u003cul\u003e\n\u003cli\u003eLocal LLM inference and AI assistants\u003c\/li\u003e\n\u003cli\u003eFine-tuning selected models\u003c\/li\u003e\n\u003cli\u003eComputer vision and deep learning\u003c\/li\u003e\n\u003cli\u003eDiffusion models and generative AI\u003c\/li\u003e\n\u003cli\u003ePyTorch, TensorFlow and CUDA development\u003c\/li\u003e\n\u003cli\u003eCAD\/CAE, rendering and visualization\u003c\/li\u003e\n\u003c\/ul\u003e\u003ch3\u003eWhy GPU memory matters\u003c\/h3\u003e\u003cp\u003e24 GB of VRAM is a practical capacity for many local LLM, computer-vision and generative-AI workflows. Whether a specific model fits also depends on precision, quantization, context length and batch size.\u003c\/p\u003e\u003cp\u003e\u003cstrong\u003eBest suited to:\u003c\/strong\u003e professional local AI development and combined AI\/engineering workflows.\u003c\/p\u003e","brand":"NVIDIA","offers":[{"title":"Default Title","offer_id":47168085721190,"sku":"900-5G147-2550-000","price":5170.0,"currency_code":"EUR","in_stock":false}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0680\/6572\/2470\/files\/nvidia-rtx-pro-4500-blackwell-32gb-front.jpg?v=1789295270"},{"product_id":"nvidia-rtx-pro-4000-blackwell-sff-retail-24-gb-gddr7-ecc-pcie-5-0-x8","title":"NVIDIA RTX PRO 4000 Blackwell SFF Retail – 24 GB GDDR7 ECC – PCIe 5.0 x8","description":"\u003cp\u003eProfessional NVIDIA RTX PRO workstation GPU with substantial GB of graphics memory for local AI, machine learning, CUDA development and technical visualization.\u003c\/p\u003e\u003ch3\u003eAI \u0026amp; ML use\u003c\/h3\u003e\u003cul\u003e\n\u003cli\u003eLocal LLM inference and AI assistants\u003c\/li\u003e\n\u003cli\u003eFine-tuning selected models\u003c\/li\u003e\n\u003cli\u003eComputer vision and deep learning\u003c\/li\u003e\n\u003cli\u003eDiffusion models and generative AI\u003c\/li\u003e\n\u003cli\u003ePyTorch, TensorFlow and CUDA development\u003c\/li\u003e\n\u003cli\u003eCAD\/CAE, rendering and visualization\u003c\/li\u003e\n\u003c\/ul\u003e\u003ch3\u003eWhy GPU memory matters\u003c\/h3\u003e\u003cp\u003e24 GB of VRAM is a practical capacity for many local LLM, computer-vision and generative-AI workflows. Whether a specific model fits also depends on precision, quantization, context length and batch size.\u003c\/p\u003e\u003cp\u003e\u003cstrong\u003eBest suited to:\u003c\/strong\u003e professional local AI development and combined AI\/engineering workflows.\u003c\/p\u003e","brand":"NVIDIA","offers":[{"title":"Default Title","offer_id":47168088866918,"sku":"900-5G195-2501-000","price":3380.0,"currency_code":"EUR","in_stock":false}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0680\/6572\/2470\/files\/nvidia-rtx-pro-4000-blackwell-sff-900-5g195-2501-000-front.jpg?v=1789295670"},{"product_id":"lenovo-thinkpad-t16g-gen-3-core-ultra-9-275hx-rtx-5090-24-gb-128-gb-1-tb","title":"Lenovo ThinkPad T16g Gen 3 – Core Ultra 9 275HX – RTX 5090 24 GB – 128 GB – 1 TB","description":"\u003cp\u003eHigh-performance mobile workstation for AI, machine learning, CUDA and engineering workloads. The ThinkPad T16g Gen 3 combines Intel Core Ultra 9 processing with NVIDIA GeForce RTX 5090 graphics and a large 128 GB memory configuration for demanding professional workflows.\u003c\/p\u003e\n\u003ch3\u003eKey specifications\u003c\/h3\u003e\n\u003cul\u003e\n\u003cli\u003eIntel Core Ultra 9 275HX, 24 cores, up to 5.4 GHz\u003c\/li\u003e\n\u003cli\u003eNVIDIA GeForce RTX 5090 with 24 GB GDDR7\u003c\/li\u003e\n\u003cli\u003e128 GB DDR5-5600 RAM (4 × 32 GB), up to 192 GB supported\u003c\/li\u003e\n\u003cli\u003e1 TB PCIe 5.0 NVMe SSD with TCG Opal Encryption 2\u003c\/li\u003e\n\u003cli\u003e16-inch WQUXGA IPS display, 3840 × 2400, 800 cd\/m², 100% DCI-P3\u003c\/li\u003e\n\u003cli\u003eWi-Fi 7, Bluetooth 5.4, NFC and 2.5 Gigabit Ethernet\u003c\/li\u003e\n\u003cli\u003e5G WWAN upgrade-ready\u003c\/li\u003e\n\u003cli\u003eWindows 11 Pro, German localization and German keyboard\u003c\/li\u003e\n\u003cli\u003e3-year Lenovo Premier Support\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003ch3\u003eAI \u0026amp; engineering use\u003c\/h3\u003e\n\u003cp\u003eSuitable for local AI inference and development, CUDA-accelerated workflows, CAD\/CAE, simulation pre\/post-processing, scientific computing, visualization and other GPU-intensive professional applications.\u003c\/p\u003e","brand":"Lenovo","offers":[{"title":"Default Title","offer_id":48962328100966,"sku":"21V5002WGE","price":9070.0,"currency_code":"EUR","in_stock":false}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0680\/6572\/2470\/files\/lenovo-thinkpad-t16g-gen3-21v50006ge-01.jpg?v=1789206137"},{"product_id":"nvidia-quantum-2-qm9700-ndr-infiniband-switch-64-port-400-gb-s-32-osfp-managed","title":"NVIDIA Quantum-2 QM9700 NDR InfiniBand Switch – 64-Port 400 Gb\/s – 32× OSFP – Managed","description":"\u003cp\u003eNVIDIA Quantum-2 QM9700 is a managed 1U NDR InfiniBand switch designed for large-scale AI, HPC and accelerated data-center fabrics. It provides 64 NDR ports through 32 OSFP cages, with up to 400 Gb\/s per port and 25.6 Tb\/s aggregate switching throughput.\u003c\/p\u003e\n\u003ch3\u003eKey specifications\u003c\/h3\u003e\n\u003cul\u003e\n\u003cli\u003eNVIDIA Quantum-2 NDR InfiniBand architecture\u003c\/li\u003e\n\u003cli\u003e64 NDR ports via 32 OSFP cages\u003c\/li\u003e\n\u003cli\u003eUp to 400 Gb\/s per port\u003c\/li\u003e\n\u003cli\u003e25.6 Tb\/s aggregate switching throughput\u003c\/li\u003e\n\u003cli\u003eManaged switch platform\u003c\/li\u003e\n\u003cli\u003e1U rack-mount form factor\u003c\/li\u003e\n\u003cli\u003ePower-to-connector (P2C) forward airflow\u003c\/li\u003e\n\u003cli\u003eNVIDIA In-Network Computing and SHARP acceleration\u003c\/li\u003e\n\u003cli\u003eDesigned for AI training clusters, HPC, supercomputing and hyperscale infrastructure\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003cp\u003e\u003cstrong\u003eManufacturer part number:\u003c\/strong\u003e 920-9B210-00FN-0M0\u003cbr\u003e\u003cstrong\u003eTD SYNNEX:\u003c\/strong\u003e 7803741\u003cbr\u003e\u003cstrong\u003eEAN:\u003c\/strong\u003e 7290108489088\u003c\/p\u003e","brand":"NVIDIA","offers":[{"title":"Default Title","offer_id":67619571662950,"sku":"920-9B210-00FN-0M0","price":37850.85,"currency_code":"EUR","in_stock":false}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0680\/6572\/2470\/files\/nvidia-quantum-2-qm9700-infiniband-switch.jpg?v=1791011848"},{"product_id":"nvidia-connectx-7-400gbe-ndr-400-infiniband-network-adapter-pcie-5-0-x16-single-port-osfp","title":"NVIDIA ConnectX-7 400GbE \/ NDR 400 InfiniBand Network Adapter – PCIe 5.0 x16 – Single-Port OSFP","description":"\u003cp\u003eNVIDIA ConnectX-7 high-performance network adapter for AI, HPC and accelerated data-center systems. This HHHL PCIe adapter supports 400Gb Ethernet and NDR 400Gb\/s InfiniBand through a single OSFP port.\u003c\/p\u003e\n\u003ch3\u003eKey specifications\u003c\/h3\u003e\n\u003cul\u003e\n\u003cli\u003e400Gb Ethernet \/ NDR 400Gb\/s InfiniBand\u003c\/li\u003e\n\u003cli\u003eSingle-port OSFP\u003c\/li\u003e\n\u003cli\u003ePCIe 5.0 x16 interface\u003c\/li\u003e\n\u003cli\u003eHalf-height, half-length (HHHL) adapter format\u003c\/li\u003e\n\u003cli\u003eSecure Boot enabled\u003c\/li\u003e\n\u003cli\u003eCrypto disabled on this SKU\u003c\/li\u003e\n\u003cli\u003eSuitable for NVIDIA Quantum-2 \/ QM9700 NDR fabrics\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003cp\u003e\u003cstrong\u003eNVIDIA SKU:\u003c\/strong\u003e 900-9X766-003N-SQ0\u003cbr\u003e\u003cstrong\u003eLegacy OPN:\u003c\/strong\u003e MCX75310AAS-NEAT\u003cbr\u003e\u003cstrong\u003eTD SYNNEX:\u003c\/strong\u003e 7793450\u003c\/p\u003e\n\u003cp\u003eFor a direct QM9700-to-host configuration, this adapter can be paired with the NVIDIA 980-9I924-00N002 2 m OSFP splitter cable.\u003c\/p\u003e","brand":"NVIDIA","offers":[{"title":"Default Title","offer_id":67619612131430,"sku":"900-9X766-003N-SQ0","price":2130.0,"currency_code":"EUR","in_stock":false}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0680\/6572\/2470\/files\/nvidia-connectx-7-mcx75310aas-neat-front.jpg?v=1791014195"},{"product_id":"nvidia-ndr-800g-to-2-400g-osfp-splitter-cable-2-m-passive-980-9i924-00n002","title":"NVIDIA NDR 800G to 2×400G OSFP Splitter Cable – 2 m – Passive – 980-9I924-00N002","description":"\u003cp\u003eNVIDIA passive copper splitter cable for high-performance NDR InfiniBand and 800G interconnects. The cable splits one 800G twin-port OSFP connection into two 400G OSFP endpoints.\u003c\/p\u003e\n\u003ch3\u003eKey specifications\u003c\/h3\u003e\n\u003cul\u003e\n\u003cli\u003e800 Gb\/s (2 × 400 Gb\/s) to 2 × 400 Gb\/s\u003c\/li\u003e\n\u003cli\u003eOSFP to 2 × OSFP\u003c\/li\u003e\n\u003cli\u003e2 metre passive copper cable\u003c\/li\u003e\n\u003cli\u003eFinned-top switch-side connector to flat-top endpoint connectors\u003c\/li\u003e\n\u003cli\u003eValidated for NVIDIA NDR InfiniBand\u003c\/li\u003e\n\u003cli\u003eSuitable for connecting a Quantum-2 QM9700 switch port to two 400G OSFP endpoints such as compatible ConnectX-7 adapters\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003cp\u003e\u003cstrong\u003eNVIDIA SKU:\u003c\/strong\u003e 980-9I924-00N002\u003cbr\u003e\u003cstrong\u003eLegacy OPN:\u003c\/strong\u003e MCP7Y00-N002\u003cbr\u003e\u003cstrong\u003eTD SYNNEX:\u003c\/strong\u003e 8169940\u003c\/p\u003e","brand":"NVIDIA","offers":[{"title":"Default Title","offer_id":67619612721254,"sku":"980-9I924-00N002","price":1040.0,"currency_code":"EUR","in_stock":false}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0680\/6572\/2470\/files\/nvidia-980-9i924-00n002-2m-osfp-splitter.jpg?v=1791013848"},{"product_id":"lenovo-thinkpad-p16-gen-3-core-ultra-9-275hx-rtx-pro-5000-blackwell-24-gb-128-gb-1-tb","title":"Lenovo ThinkPad P16 Gen 3 – Core Ultra 9 275HX – RTX PRO 5000 Blackwell 24 GB – 128 GB – 1 TB","description":"\u003cp\u003eFlagship Lenovo ThinkPad P16 Gen 3 mobile AI workstation for demanding AI, machine-learning, CAD\/CAE, rendering and engineering workloads. This German TopSeller configuration combines Intel Core Ultra 9 275HX processing with NVIDIA RTX PRO 5000 Blackwell graphics, 24 GB GDDR7 VRAM, 128 GB DDR5 memory and a 1 TB PCIe 5.0 NVMe SSD.\u003c\/p\u003e\n\u003ch3\u003eKey specifications\u003c\/h3\u003e\n\u003cul\u003e\n\u003cli\u003eIntel Core Ultra 9 275HX, 24 cores, up to 5.4 GHz, 36 MB cache\u003c\/li\u003e\n\u003cli\u003eNVIDIA RTX PRO 5000 Blackwell, 24 GB GDDR7, up to 1824 TOPS\u003c\/li\u003e\n\u003cli\u003e128 GB DDR5-5600 RAM (4 × 32 GB), up to 192 GB supported\u003c\/li\u003e\n\u003cli\u003e1 TB M.2 2280 PCIe 5.0 x4 NVMe SSD with TCG Opal 2.0\u003c\/li\u003e\n\u003cli\u003e16-inch WQUXGA IPS display, 3840 × 2400, 800 nits, 100% DCI-P3, HDR400, Dolby Vision\u003c\/li\u003e\n\u003cli\u003eIntel Wi-Fi 7 BE200, Bluetooth 5.4 and 2.5 Gigabit Ethernet\u003c\/li\u003e\n\u003cli\u003e5G WWAN upgradeable\u003c\/li\u003e\n\u003cli\u003e2 × Thunderbolt 5, Thunderbolt 4, HDMI with 8K support\u003c\/li\u003e\n\u003cli\u003e5 MP IR camera, fingerprint reader and discrete TPM 2.0\u003c\/li\u003e\n\u003cli\u003e99.9 Wh battery, 180 W USB-C GaN power adapter\u003c\/li\u003e\n\u003cli\u003eGerman backlit keyboard with numeric keypad\u003c\/li\u003e\n\u003cli\u003eWindows 11 Pro, German\u003c\/li\u003e\n\u003cli\u003e3 years Lenovo Premier Support plus 0.5 t CO2 offset\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003cp\u003e\u003cstrong\u003eManufacturer part number:\u003c\/strong\u003e 21RQ005YGE\u003cbr\u003e\u003cstrong\u003eTD SYNNEX:\u003c\/strong\u003e 13276534\u003cbr\u003e\u003cstrong\u003eEAN:\u003c\/strong\u003e 0828143024819\u003c\/p\u003e","brand":"Lenovo","offers":[{"title":"Default Title","offer_id":67716078960742,"sku":"21RQ005YGE","price":9530.0,"currency_code":"EUR","in_stock":false}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0680\/6572\/2470\/files\/lenovo-thinkpad-p16-gen3-21rq005yge-hero.jpg?v=1792101030"}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0680\/6572\/2470\/collections\/minimialst-and-technical.png?v=1789323276","url":"https:\/\/www.likua.de\/collections\/hardware.oembed?page=2","provider":"LIKUA","version":"1.0","type":"link"}