Skip to main content Skip to search Skip to main navigation
Menu

H-Series (Hopper) ~ PCIe 4.0

Filter products

3 Products
NVIDIA H100 NVL  94GB  PCIe 5.0 x16 NVIDIA H100 NVL  94GB  PCIe 5.0 x16
Available from 03.09.2026
Prod. no.: A17564
NVIDIA H100 NVL 94GB PCIe 5.0 x16

The NVIDIA H100 NVL is a high-performance data center GPU based on the NVIDIA Hopper architecture and was specifically designed for demanding AI, machine learning, and high-performance computing workloads. Its large HBM3 memory with high memory bandwidth enables the efficient processing of large AI models and massive datasets. The H100 NVL is particularly well-suited for large language models (LLMs), generative AI, deep learning, scientific computing, and other GPU-accelerated applications. Compatible GPUs can be interconnected via NVIDIA NVLink for particularly memory- and compute-intensive workloads. The passive cooling is designed for appropriately configured server and data center systems.Package Contents:• NVIDIA H100 NVL 94GB PCIe

Regular price: €38,999.00
NVIDIA H100 80GB PCIe 4.0 NVIDIA H100 80GB PCIe 4.0
Out of stock
Prod. no.: A16852
NVIDIA H100 80GB PCIe 4.0

GPU name: GH100 Architecture: Hopper Memory size: 80 GB Memory type: HBM2e Memory bus: 5120 bit Memory bandwidth: 1280 GB/s Supported APIs: OpenCL / DirectCompute / OpenACC / CUDA Shader units: 7,296 Tensor cores: 456 FP32 (Float) Power: 24.08 TFLOPS FP64 (double) Power: 12.04 TFLOPS Power consumption: 350 Watt Thermal solution: Passive

Regular price: €34,999.00
NVIDIA H200 NVL 141GB HBM3e PCIe 5.0 NVIDIA H200 NVL 141GB HBM3e PCIe 5.0
Out of stock
Prod. no.: A18060
NVIDIA H200 NVL 141GB HBM3e PCIe 5.0

The NVIDIA H200 NVL 141GB is a high-end computing accelerator card designed for demanding AI, HPC, and data center workloads. Based on the NVIDIA Hopper architecture and equipped with 141 GB of HBM3e ECC memory, it offers enormous memory bandwidth and computing power for large language models (LLMs), generative AI, machine learning, scientific simulations, and data-intensive applications.The exceptionally large HBM3e memory enables the processing of extensive AI models and datasets directly on the GPU. Multiple H200 NVL GPUs can be interconnected via NVLink for appropriately scalable workloads. The passive cooling is designed for use in dedicated server and workstation systems with sufficient system airflow.The H200 NVL has no display outputs and is designed as a compute accelerator for GPU computing. It is particularly well-suited for professional systems where maximum AI computing power, large GPU memory, and high memory bandwidth are paramount.Package Contents:• NVIDIA H200 NVL 141GB PCIe

Regular price: €27,273.61