加速卡Accelerators
NVIDIA H100 SXMNVIDIA H100 SXM
回答数据中心训练与推理卡的显存容量、带宽、功耗与互联口径。不回答实际训练吞吐或单位算力价格:这里记的是官方规格表数字,且带稀疏标注,实测吞吐随框架、精度与模型结构变化。Answers the memory capacity, bandwidth, power and interconnect figures of a data-centre training and inference card. It does not answer real training throughput or price per unit of compute: these are vendor spec-sheet numbers with sparsity, and measured throughput varies with framework, precision and model shape.
关键参数Key specs
- memory
- 80 GB HBM3
- memory bandwidth
- 3.35 TB/s
- max TDP
- up to 700 W (configurable)
- form factor
- SXM
- interconnect
- NVLink 900 GB/s; PCIe Gen5 128 GB/s
- FP8 tensor, sparsity
- 3,958 TFLOPS
- MIG
- up to 7 instances at 10 GB each
官方原文怎么说What the source says
GPU Memory 80GB 94GB
GPU Memory Bandwidth 3.35TB/s 3.9TB/s
Max Thermal Design Power (TDP) Up to 700W (configurable) 350-400W (configurable)
Interconnect NVIDIA NVLink™: 900GB/s PCIe Gen5: 128GB/s
资料与许可Source and licence
- 资料时间Source checked
- 2026-11-07
- 许可 / 类型Licence / kind
- official-spec-page
- 许可说明Licence evidence
- https://www.nvidia.com/en-us/data-center/h100/
- 类目Category
- 加速卡Accelerators