Nvidia Blackwell architecture
GPTKB entity
Statements (25)
| Predicate | Object |
|---|---|
| gptkbp:instanceOf |
gptkb:graphics_card
|
| gptkbp:announced |
2024
|
| gptkbp:developedBy |
gptkb:Nvidia
|
| https://www.w3.org/2000/01/rdf-schema#label |
Nvidia Blackwell architecture
|
| gptkbp:notableFeature |
gptkb:Transformer_Engine
Improved energy efficiency Enhanced AI inference performance Multi-die GPU design Second-generation NVLink Switch System |
| gptkbp:predecessor |
gptkb:Nvidia_Hopper_architecture
|
| gptkbp:processNode |
gptkb:TSMC_4NP
|
| gptkbp:successor |
Nvidia Rubin architecture
|
| gptkbp:supports |
gptkb:HBM3E
gptkb:NVLink_5.0 gptkb:PCIe_Gen5 BF16 FP16 FP8 TensorRT-LLM |
| gptkbp:targetMarket |
gptkb:artificial_intelligence
High-performance computing |
| gptkbp:usedIn |
Nvidia B100
Nvidia B200 |
| gptkbp:bfsParent |
gptkb:Nvidia_Hopper_architecture
|
| gptkbp:bfsLayer |
7
|