GPU Wiki

Compare NVIDIA 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.

Swap Clear
Row order:
Muted cells indicate unavailable, not applicable, or not directly comparable specifications.
Spec Tesla K20s
Overview
Product name NVIDIA Tesla K20s
Architecture Kepler
Generation / family Tesla K Series
GPU codename GK110
Launch date 2013-02-18
Market segment Data Center in Tesla K Series
Board type / entry type Data Center GPU
GPU
Process node TSMC 28 nm
Die size 561 mm²
Transistor count 7.08 billion
GPU chip GK110
SM / multiprocessor count 0
CUDA Cores 2496
ROPs 40
TMUs 208
Clocks
Base / Core clock 575 MHz
Boost clock 575 MHz
Memory clock 1,300 MHz
Memory
Memory size 5 GB
Memory type GDDR5
Memory bus width 320-bit
Memory bandwidth 208 GB/s
Bus / Interface
Bus interface PCIe 2.0 x16
AGP / PCI / PCIe distinction PCIe
Power
TDP / TBP / board power 225 W
Recommended PSU 550 W
Slot width Dual-slot, 267 mm length
Features
DirectX support DirectX 11.0
OpenGL support OpenGL 4.6
Vulkan support Vulkan 1.2
CUDA support 3.5
Shader Model 5.1
Performance / Benchmarks
FP32 throughput 2.87 TFLOPS

© 2026 NVIDIA Club. Independent GPU history, guides, and hardware knowledge.

NVIDIA Club is an independent enthusiast and technical knowledge site. It is not affiliated with, endorsed by, or sponsored by NVIDIA Corporation.