Nvidia GH200 Grace Hopper Superchip
GPTKB entity
Statements (33)
| Predicate | Object |
|---|---|
| gptkbp:instanceOf |
gptkb:microprocessor
|
| gptkbp:announced |
2023
|
| gptkbp:architecture |
gptkb:Grace_Hopper
|
| gptkbp:capacity |
up to 480 GB HBM3e
|
| gptkbp:connects |
gptkb:NVLink-C2C
|
| gptkbp:GPU |
gptkb:Nvidia_Hopper_GPU
|
| https://www.w3.org/2000/01/rdf-schema#label |
Nvidia GH200 Grace Hopper Superchip
|
| gptkbp:manufacturer |
gptkb:Nvidia
|
| gptkbp:memoryType |
gptkb:HBM3
|
| gptkbp:namedAfter |
gptkb:Grace_Hopper
gptkb:Grace_CPU gptkb:Hopper_GPU |
| gptkbp:processNode |
gptkb:TSMC_4N
|
| gptkbp:processor |
gptkb:Nvidia_Grace_CPU
|
| gptkbp:supports |
gptkb:Confidential_Computing
gptkb:PCIe_Gen5 gptkb:Nvidia_HPC_SDK gptkb:Nvidia_cuQuantum gptkb:Nvidia_AI_Enterprise gptkb:Nvidia_CUDA gptkb:Nvidia_NVLink gptkb:Nvidia_NVSwitch gptkb:Nvidia_Omniverse gptkb:Tensor_Cores FP16 precision FP8 precision |
| gptkbp:targetMarket |
gptkb:artificial_intelligence
high-performance computing |
| gptkbp:used_in |
gptkb:Nvidia_DGX_GH200_system
cloud computing supercomputers |
| gptkbp:bfsParent |
gptkb:Nvidia_Hopper
|
| gptkbp:bfsLayer |
6
|