|
gptkbp:instance_of
|
gptkb:Graphics_Processing_Unit
|
|
gptkbp:application
|
gptkb:machine_learning
deep learning
graphics rendering
AI inference
|
|
gptkbp:architecture
|
gptkb:Turing
|
|
gptkbp:competes_with
|
gptkb:AMD_Radeon_Instinct_MI_series
gptkb:Google_TPU
|
|
gptkbp:designed_by
|
gptkb:NVIDIA
|
|
gptkbp:form_factor
|
gptkb:PCIe
Single-slot
|
|
https://www.w3.org/2000/01/rdf-schema#label
|
NVIDIA T4 GPUs
|
|
gptkbp:manufacturing_process
|
12 nm
|
|
gptkbp:market_segment
|
gptkb:education
gptkb:Telecommunications
gptkb:research
gptkb:government
gptkb:financial_services
healthcare
automotive
enterprise
retail
media and entertainment
|
|
gptkbp:number_of_cores
|
320
2560
|
|
gptkbp:performance
|
130 TOPS
65 TFLOPS
8.1 TFLOPS
|
|
gptkbp:power_consumption
|
70 W
|
|
gptkbp:predecessor
|
gptkb:NVIDIA_P4
|
|
gptkbp:ram
|
16 GB GDDR6
320 GB/s
|
|
gptkbp:release_date
|
gptkb:2018
|
|
gptkbp:release_notes
|
NVIDIA T4 GPU Architecture Whitepaper
|
|
gptkbp:resolution
|
7680 x 4320
|
|
gptkbp:successor
|
gptkb:NVIDIA_A100
|
|
gptkbp:supports
|
gptkb:NVIDIA_v_GPU
gptkb:NVIDIA_NVLink
gptkb:Tensor_RT
gptkb:CUDA
gptkb:NVIDIA_Deep_Learning_SDK
|
|
gptkbp:thermal_design_power
|
70 W
|
|
gptkbp:used_in
|
gptkb:cloud_computing
data centers
high-performance computing
|
|
gptkbp:video_output
|
gptkb:4
|
|
gptkbp:bfsParent
|
gptkb:A100_Tensor_Core_GPU
|
|
gptkbp:bfsLayer
|
5
|