{{ 'fb_in_app_browser_popup.desc' | translate }} {{ 'fb_in_app_browser_popup.copy_link' | translate }}
{{ 'in_app_browser_popup.desc' | translate }}
The NVIDIA H800-80G is a high-performance GPU tailored for AI and machine learning applications in enterprise data centers. Designed by NVIDIA, this GPU offers powerful computational capabilities, catering to engineers and IT professionals seeking to optimize their AI workloads efficiently.
Not enough stock.
Your item was not added to your cart.
Not enough stock.
Please adjust your quantity.
{{'products.quick_cart.out_of_number_hint'| translate}}
{{'product.preorder_limit.hint'| translate}}
Limit {{ product.max_order_quantity }} per order.
Only {{ quantityOfStock }} item(s) left.
The NVIDIA H800-80G is a high-performance GPU tailored for AI and machine learning applications in enterprise data centers. Designed by NVIDIA, this GPU offers powerful computational capabilities, catering to engineers and IT professionals seeking to optimize their AI workloads efficiently. The H800-80G is part of the NVIDIA H800 GPU series and features 80GB of high-bandwidth memory (HBM2), enabling seamless performance for large-scale AI tasks.
Product | NVIDIA H800 | NVIDIA A100 | NVIDIA V100 | AMD MI100 | Intel Habana Gaudi | Google TPU v4 |
---|---|---|---|---|---|---|
Performance (TFLOPS) | 80 | 70 | 50 | 100 | 100 | 80 |
Memory Type | HBM2 | HBM2 | HBM2 | HBM2e | HBM | SRAM |
Memory Capacity (GB) | 80 | 40 | 32 | 32 | 32 | 16 |
PCIe Support | Gen 4 | Gen 4 | Gen 3 | Gen 4 | Gen 4 | Gen 3 |
Form Factor | PCIe | PCIe | PCIe | PCIe | OCP | PCIe |
Model | Description |
---|---|
NVIDIA NVLink Bridge | High-speed GPU interconnect |
NVIDIA TensorRT | Deep learning inference optimizer |
CUDA Toolkit | Parallel computing software platform |
Nvidia H800 80G Specifications | |
GPU Architecture | Ampere |
CUDA Cores | 10,240 |
Tensor Cores | 320 |
Memory Size | 80 GB |
Memory Type | HBM2e |
Memory Bandwidth | 2 TB/s |
Peak FP16 Performance | 134.4 TFLOPS |
Peak FP32 Performance | 67.2 TFLOPS |
Peak INT8 Performance | 1,072 TOPS |
NVLink Support | Yes, up to 12 links |
PCI Express | Gen 4 |
Form Factor | PCIe |
Power Consumption | 500W |
Thermal Solution | Active/Passive |
ECC Memory | Yes |
Multi-GPU Technology | NVLink |
Maximum GPU Temperature | 85°C |
Operating System Support | Windows, Linux |
API Support | DirectX 12, OpenGL 4.6, Vulkan, CUDA, OpenCL |