Acerca del producto
NVIDIA® H100 NVL supercharges large language model inference in mainstream PCIe-based server systems. With increased raw performance, bigger, faster HBM3 memory, and NVIDIA NVLink™ connectivity via bridges, mainstream systems with H100 NVL outperform NVIDIA A100 Tensor Core systems by up to 5X on Llama 2 70B. H100 NVL also includes NVIDIA Enterprise software.
- FP64 -30 teraFLOPS
- FP64 Tensor Core - 60 teraFLOPS
- FP32 - 60 teraFLOPS
- TF32 Tensor Core - 835 teraFLOPS
- BFLOAT16 Tensor Core - 1,671 teraFLOPS
- FP16 Tensor Core - 1,671 teraFLOPS
- FP8 Tensor Core - 3,341 teraFLOPS
- INT8 Tensor Core - 3,341 teraFLOPS
- GPU Memory - 94GB
- GPU Memory Bandwidth - 3.9TB/s
- Decoders 7 NVDEC / 7 JPEG
- Max Thermal Design Power (TDP)- 350-400W (configurable)
- Form Factor - PCIe dual-slot air-cooled
- Interconnect - NVIDIA NVLINK: 600GB/s
- PCIe Gen5 - 128GB/s
- Server Options Partner and NVIDIA
- NVIDIA Enterprise Included
Especificaciones técnicas
| Interfaces/Ports | |
| Host Interface | PCI Express |
| Memory | |
| Standard Memory | 94 GB |
| Memory Technology | HBM3 |
| Miscellaneous | |
| Package Contents |
|
| Platform Supported | PC |
| Processor & Chipset | |
| Chipset Manufacturer | NVIDIA |
| Chipset Model | H100 NVL |
| Technical Information | |
| Multi-GPU Technology | NVLink |
| AI-powered | Yes |
| Physical Characteristics | |
| Form Factor | Plug-in Card |
| Cooler Type | Fan Cooler |
| Slot Space Required | Dual |
| General Information | |
| Product Type | Graphic Card |
| Manufacturer Part Number | NVH100NVLTCGPUKIT |
| Manufacturer Website Address | http://www.pny.com |
| Manufacturer | PNY Technologies |
| Product Name | NVIDIA H100 NVL Graphic Card |
| Brand Name | PNY |
Comparación de productos
Garantía del producto
Garantía estándar del fabricante según corresponda
Política de devoluciones
Garantía de precio
!00% siempre estamos por debajo del precio del mercado.
Explora colecciones de marca
Colecciones
Nuestros servicios
