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.

Add a third or fourth GPU
Spec A10 PCIe
Overview
Product name NVIDIA A10 PCIe
Architecture Ampere
Generation / family NVIDIA A-Series
GPU codename GA102
Launch date 2021-04-12
Market segment Data Center in NVIDIA A-Series
Board type / entry type Data Center GPU
GPU
Process node Samsung 8 nm
Die size 628 mm²
Transistor count 28.3 billion
GPU chip GA102
SM / multiprocessor count 72
CUDA Cores 9216
Tensor Cores 288
RT Cores 72
ROPs 96
TMUs 288
Clocks
Base / Core clock 885 MHz
Boost clock 1,695 MHz
Memory clock 1,563 MHz
Memory
Memory size 24 GB
Memory type GDDR6
Memory bus width 384-bit
Memory bandwidth 600.2 GB/s
Bus / Interface
Bus interface PCIe 4.0 x16
AGP / PCI / PCIe distinction PCIe
Power
TDP / TBP / board power 150 W
Recommended PSU 450 W
Power connectors 1x PCIe 8-pin
Slot width Single-slot, 267 mm length
Features
DirectX support DirectX 12.2
OpenGL support OpenGL 4.6
Vulkan support Vulkan 1.4
CUDA support 8.6
Shader Model 6.8
Ray tracing support Hardware RT units present
Performance / Benchmarks
FP32 throughput 31.24 TFLOPS
FP16 throughput 31.24 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.