Statements (24)
| Predicate | Object |
|---|---|
| gptkbp:instanceOf |
gptkb:Neural_Engine
|
| gptkbp:announced |
gptkb:Google
2023 |
| gptkbp:architecture |
ASIC
|
| gptkbp:availableOn |
gptkb:Google_Cloud_Vertex_AI
|
| gptkbp:cloudBased |
gptkb:Google_Cloud
|
| gptkbp:connects |
high-speed interconnect
|
| gptkbp:designedFor |
gptkb:artificial_intelligence
gptkb:machine_learning |
| gptkbp:energyEfficiency |
improved over previous generation
|
| gptkbp:manufacturer |
gptkb:Google
|
| gptkbp:maximumSpeed |
100 teraFLOPS
|
| gptkbp:memoryPerChip |
16 GB
|
| gptkbp:predecessor |
gptkb:TPU_v4
|
| gptkbp:size |
multi-chip pod configurations
|
| gptkbp:supports |
training
inference |
| gptkbp:uses |
computer vision
recommendation systems large language models |
| gptkbp:bfsParent |
gptkb:TPU_v4
gptkb:TPU_Pods |
| gptkbp:bfsLayer |
8
|
| https://www.w3.org/2000/01/rdf-schema#label |
TPU v5e
|