Statements (23)
| Predicate | Object |
|---|---|
| gptkbp:instanceOf |
CPU instruction set extension
|
| gptkbp:alsoKnownAs |
gptkb:Vector_Neural_Network_Instructions
|
| gptkbp:architecture |
x86
|
| gptkbp:category |
SIMD instruction set
|
| gptkbp:developedBy |
gptkb:Intel
|
| gptkbp:enables |
faster AI inference
reduced memory bandwidth usage |
| gptkbp:features |
gptkb:Intel_Xeon_Scalable_processors
|
| https://www.w3.org/2000/01/rdf-schema#label |
Deep Learning Boost (VNNI)
|
| gptkbp:improves |
inference performance
|
| gptkbp:includedIn |
gptkb:Intel_Ice_Lake
gptkb:Intel_Tiger_Lake gptkb:Intel_Cooper_Lake |
| gptkbp:introducedIn |
gptkb:Intel_Cascade_Lake
|
| gptkbp:purpose |
accelerate deep learning workloads
|
| gptkbp:relatedTo |
gptkb:Intel_Deep_Learning_Boost
gptkb:AVX-512 |
| gptkbp:supports |
8-bit integer operations
16-bit integer operations |
| gptkbp:usedFor |
convolutional neural networks
matrix multiplication |
| gptkbp:bfsParent |
gptkb:Cascade_Lake
|
| gptkbp:bfsLayer |
5
|