NVIDIA H200 Tensor Core GPU
- Product: NVIDIA H200 Tensor Core Graphics Card
- GPU Memory: 141GB
- Memory Bandwidth: Up to 4.8TB/s
- Tensor Performance: Up to 3,958 TFLOPS
- Interconnect: NVIDIA NVLink & PCIe Gen5
- Power Consumption: Up to 700W
- Form Factor: SXM / PCIe Dual-Slot
- Warranty: 1 Year Assured Warranty
- Support: 24/7 Technical Support
- Use Cases: AI Training, Large Language Models (LLMs), Deep Learning, HPC Workloads, Generative AI, Data Center Computing
Are you experiencing slow training of AI, a lack of sufficient memory on your current graphics card, or performance challenges while running LLM or HPC workloads on the system? In order to reduce AI infrastructure bottlenecks, the NVIDIA H200 Tensor Core GPU is designed for generative AI, scientific computing, and enterprise data center workloads. The NVIDIA Hopper architecture powers the H200, providing up to 141GB of HBM3e memory, 4.8TB/s of bandwidth, 4th Generation Tensor Cores, and FP8/FP16 precision support. Features like NVLink connectivity, CUDA parallel processing, and PCIe 5.0 compatibility help to speed up times across multiple GPUs. NVIDIA GPU H200 is available in both SXM and PCIe configurations for enterprise AI servers and HPC clusters. ServerBasket UAE offers 24/7 technical assistance with a one-year warranty on all NVIDIA graphics cards.
Technical Specifications
| FP64 Performance | 34 TFLOPS |
|---|---|
| FP64 Tensor Core | 67 TFLOPS |
| FP32 Performance | 67 TFLOPS |
| TF32 Tensor Core | 989 TFLOPS |
| BFLOAT16 Tensor Core | 1,979 TFLOPS |
| FP16 Tensor Core | 1,979 TFLOPS |
| FP8 Tensor Core | 3,958 TFLOPS |
| INT8 Tensor Core | 3,958 TFLOPS |
| GPU Memory | 141GB |
| Memory Bandwidth | 4.8 TB/s |
| Decoders | 7 NVDEC, 7 JPEG |
| Confidential Computing | Supported |
| Max Power Consumption | Up to 700W |
| Multi-Instance GPU | Up to 7 MIGs @ 18GB |
| Form Factor | SXM |
| Interconnect | NVLink 900GB/s, PCIe Gen5 128GB/s |
| Server Support | NVIDIA HGX H200 Systems |
| AI Enterprise Support | Included |
| Ideal Use Cases | AI Training, HPC, LLMs, Deep Learning |
Technical Specifications
| FP64 Performance | 30 TFLOPS |
|---|---|
| FP64 Tensor Core | 60 TFLOPS |
| FP32 Performance | 60 TFLOPS |
| TF32 Tensor Core | 835 TFLOPS |
| BFLOAT16 Tensor Core | 1,671 TFLOPS |
| FP16 Tensor Core | 1,671 TFLOPS |
| FP8 Tensor Core | 3,341 TFLOPS |
| INT8 Tensor Core | 3,341 TFLOPS |
| GPU Memory | 141GB |
| Memory Bandwidth | 4.8 TB/s |
| Decoders | 7 NVDEC, 7 JPEG |
| Confidential Computing | Supported |
| Max Power Consumption | Up to 600W |
| Multi-Instance GPU | Up to 7 MIGs @ 16.5GB |
| Form Factor | PCIe Dual-Slot Air-Cooled |
| Interconnect | NVLink Bridge 900GB/s, PCIe Gen5 128GB/s |
| Server Support | NVIDIA MGX H200 NVL Systems |
| AI Enterprise Support | Included |
| Ideal Use Cases | AI Inference, Enterprise AI, Data Centers |
No FAQ Found
Reviews
There are no reviews yet.