NVIDIA L40S vs NVIDIA Tesla K20m
GPU · Specs Compare
Graphics Card
Release Date
Oct 2022
Jan 2013
Generation
Server Ada
Tesla
Type
Professional
Professional
Bus Interface
PCIe 4.0 x16
PCIe 2.0 x16
Clock Speeds
Base Clock
1065 MHz
—
Boost Clock
2520 MHz
—
Memory Clock
2250 MHz
1300 MHz
Memory
Memory Size
48GB
5GB
Memory Type
GDDR6
GDDR5
Memory Bus
384bit
320bit
Bandwidth
864.0GB/s
208.0GB/s
Render Config
SM Count
142
—
Shading Units
18176
2496
TMUs
568
208
ROPs
192
40
Tensor Cores
568
—
RT Cores
142
—
L1 Cache
128 KB (per SM)
16 KB (per SMX)
L2 Cache
48 MB
1280 KB
Theoretical Performance
Pixel Rate
483.8 GPixel/s
36.71 GPixel/s
Texture Rate
1431.4 GTexel/s
146.8 GTexel/s
FP16 (half)
91.61 TFLOPS
—
FP32 (float)
91.61 TFLOPS
3.524 TFLOPS
FP64 (double)
1431.4 GFLOPS
1175 GFLOPS
Graphics Processor
GPU Name
AD102
GK110
GPU Variant
AD102-896-A1
—
Architecture
Ada Lovelace
Kepler
Foundry
TSMC
TSMC
Process Size
5 nm
28 nm
Transistors
76.3 billion
7.08 billion
Die Size
609 mm²
561 mm²
Board Design
TDP
350W
225W
Suggested PSU
750 W
550 W
Outputs
4x DisplayPort 1.4a
No outputs
Power Connectors
1x 16-pin
1x 6-pin + 1x 8-pin
Graphics Features
DirectX
12 Ultimate (12_2)
12 (11_0)
OpenGL
4.6
4.6
OpenCL
3.0
3.0
Vulkan
1.4
1.1
CUDA
8.9
3.5
Shader Model
6.9
5.1
Benchmark
Blender
9,239
—
▲ Specs Compare
Data sourced from official specifications and benchmark tests.