|
gptkbp:instanceOf
|
hardware component
|
|
gptkbp:accelerate
|
matrix multiplication
|
|
gptkbp:architecture
|
gptkb:Ada_Lovelace
gptkb:Volta
gptkb:Hopper
gptkb:Ampere
gptkb:Turing
|
|
gptkbp:developedBy
|
gptkb:NVIDIA
|
|
gptkbp:feature
|
gptkb:DLSS
energy efficiency
parallel processing
AI denoising
high throughput
AI super resolution
AI-powered ray tracing
|
|
gptkbp:firstAppearance
|
gptkb:NVIDIA_Tesla_V100
|
|
https://www.w3.org/2000/01/rdf-schema#label
|
NVIDIA Tensor Cores
|
|
gptkbp:introducedIn
|
gptkb:Volta_architecture
|
|
gptkbp:optimizedFor
|
deep learning
|
|
gptkbp:supportedBy
|
dot product
convolution
mixed precision computing
GEMM
matrix multiply-accumulate
|
|
gptkbp:supportLibrary
|
gptkb:CUDA
gptkb:TensorFlow
gptkb:TensorRT
gptkb:cuDNN
gptkb:MXNet
gptkb:PyTorch
gptkb:ONNX
|
|
gptkbp:supportPrecision
|
gptkb:BFLOAT16
FP16
FP8
INT8
FP64
TF32
|
|
gptkbp:usedFor
|
scientific computing
graphics rendering
AI inference
AI training
high performance computing
|
|
gptkbp:usedIn
|
gptkb:NVIDIA_A100
gptkb:NVIDIA_GPUs
gptkb:NVIDIA_H100
gptkb:NVIDIA_L40
gptkb:NVIDIA_RTX_30_Series
gptkb:NVIDIA_RTX_40_Series
gptkb:NVIDIA_Titan_V
gptkb:NVIDIA_Quadro_RTX
NVIDIA RTX 20 Series
|
|
gptkbp:bfsParent
|
gptkb:Rapid_Packed_Math
|
|
gptkbp:bfsLayer
|
7
|