Brand: NVIDIA | Category: GPUs
SKU: NV-H200-141GB | Part #: H200-141GB | MPN: 900-21010-0020-000
Contact for Pricing — Request a Quote
The NVIDIA H200 Tensor Core GPU 141GB HBM3e from NVIDIA is enterprise-grade GPUs hardware built for data centers. It features 141GB capacity, PCIe 4.0 x16 form factor. NVIDIA H200 Tensor Core GPU with 141GB HBM3e memory delivering 1.9x inference improvement over H100 for large language models. Available in brand new condition. NVIDIA gpus products are trusted by enterprises and data centers worldwide. Available from Omnixon Global. Contact our sales team to request a quote.
| Brand | NVIDIA |
| Category | GPUs |
| SKU | NV-H200-141GB |
| Part Number | H200-141GB |
| MPN | 900-21010-0020-000 |
| Condition | New |
| Capacity | 141GB |
| Form Factor | PCIe 4.0 x16 |
| GPU Architecture | NVIDIA Hopper |
| RAID Support | H200 |
| GPU Support | NVIDIA H200 |
| AI Optimized | Yes |
| Max Memory | 141GB HBM3e |
| Memory Bandwidth | 3.2 TB/s |
| Tensor Cores | 4th Generation Tensor Cores |
| CUDA Cores | Up to 16,384 CUDA cores |
| Interconnect | NVIDIA NVLink 4th Gen |
| Power Consumption | Approx. 350W |
| Form Factor | PCIe 4.0 x16 |
| Supported APIs | CUDA, OpenCL, Vulkan, DirectX |
| Cooling Solution | Active cooling with dual fans |
The NVIDIA H200 Tensor Core GPU 141GB HBM3e accelerates AI/ML training, inference, scientific HPC, and virtualization (vGPU) workloads. Typical deployments include LLM training clusters, computer-vision pipelines, financial risk modeling, and rendering farms.
Key specifications for the NVIDIA H200 Tensor Core GPU 141GB HBM3e: 3 Years support; new condition; gpu architecture NVIDIA Hopper; memory capacity 141GB HBM3e; memory bandwidth 3.2 TB/s; tensor cores 4th Generation Tensor Cores; cuda cores Up to 16,384 CUDA cores; interconnect NVIDIA NVLink 4th Gen. Manufacturer part number H200-141GB. For the full datasheet with electrical, environmental, and compliance details, contact our pre-sales engineering team.
The NVIDIA H200 Tensor Core GPU 141GB HBM3e requires a PCIe Gen4 or Gen5 x16 slot, server power adequate for the card's TDP, and CUDA/ROCm driver support in your hypervisor or bare-metal OS. Sales engineering will confirm chassis fit (1U/2U/4U), PCIe lane count, and PSU headroom before quoting.