GPU Wiki

Compare GPUs

Select up to four GPU Wiki entries to compare matching specifications row by row. The comparison uses the site database and keeps generation-specific specs visibly separate.

Popular comparisons

Add a third or fourth GPU
Spec Tesla PG503-216
Overview
Product name NVIDIA Tesla PG503-216
Architecture Volta
Generation / family Tesla P Series
GPU codename GV100
Launch date 2019-11-26
Market segment Data Center in Tesla P Series
Board type / entry type Data Center GPU
GPU
Process node TSMC 12 nm
Die size 815 mm²
Transistor count 21.1 billion
GPU chip GV100
SM / multiprocessor count 80
CUDA Cores 4608
Tensor Cores 640
ROPs 128
TMUs 288
Clocks
Base / Core clock 1,312 MHz
Boost clock 1,530 MHz
Memory clock 1,106 MHz
Memory
Memory size 32 GB
Memory type HBM2
Memory bus width 4,096-bit
Memory bandwidth 1,130 GB/s
Bus / Interface
Bus interface PCIe 3.0 x16
AGP / PCI / PCIe distinction PCIe
Power
TDP / TBP / board power 250 W
Recommended PSU 600 W
Power connectors None
Slot width Dual-slot
Features
DirectX support DirectX 12.1
OpenGL support OpenGL 4.6
Vulkan support Vulkan 1.4
CUDA support 7.0
Shader Model 6.8
Performance / Benchmarks
FP32 throughput 14.1 TFLOPS
FP16 throughput 28.2 TFLOPS
Muted cells indicate unavailable, not applicable, or not directly comparable specifications.

Frame rates compare runs of the same game at the same quality preset on one test system. DLAA counts as native resolution, because it applies anti-aliasing at full render resolution rather than upscaling; DLSS and FSR upscaling are never counted as native. Where a capture named ray tracing without a readable on or off value, it is read as on.