6 GB
GDDR5732 MHz
Boost Clock235 W
TDP28 nm
ProcessShare on Social Media
Share Tesla K20X specifications with othersTechnical Specifications
Core Specifications
Architecture:
Kepler
Process Size:
28 nm
Base Clock:
732 MHz
Boost Clock:
732 MHz
Shader Units:
2688
TMUs:
224
ROPs:
48
Memory & Performance
Memory Size:
6 GB
Memory Type:
GDDR5
Memory Bandwidth:
249.6 GB/s
Bus Interface:
PCIe 3.0 x16
TDP:
235 W
Suggested PSU:
550 W
Cache & Die Information
L1 Cache:
16 KB
L2 Cache:
1.5 MB
Die Size:
561 mm²
Transistors:
7.1 billion
Transistor Density:
12.6 million/mm²
Physical Specifications
Board Length:
267 mm
Power Connectors:
1x 6-pin + 1x 8-pin
Quick Facts
Performance
Pixel Fill Rate
41.0 GPixel/s
Texture Fill Rate
164.0 GTexel/s
FP32 Performance
3.9 TFLOPS
CUDA Toolkit Compatibility
Cuda Compute
3.5
sm_35
Architecture
Kepler
Supported CUDA Toolkits
5.0
5.5
6.0
6.5
7.0
7.5
8.0
9.0
9.1
9.2
Note: Support dropped in CUDA 10.0
Kepler architecture (GK110/GK208) - Support dropped in CUDA 10.0