NVIDIA Tesla K40s vs NVIDIA L40S
GPU Specs Compare
Graphics Card
Release Date
Nov 2013
Oct 2022
Generation
Tesla Kepler
Server Ada
Type
Professional
Professional
Bus Interface
PCIe 3.0 x16
PCIe 4.0 x16
Clock Speeds
Base Clock
706 MHz
1065 MHz
Boost Clock
706 MHz
2520 MHz
Memory Clock
1502 MHz
2250 MHz
Memory
Memory Size
12GB
48GB
Memory Type
GDDR5
GDDR6
Memory Bus
384bit
384bit
Bandwidth
288.4GB/s
864.0GB/s
Render Config
Shading Units
2880
18176
TMUs
240
568
ROPs
48
192
L1 Cache
16 KB (per SMX)
128 KB (per SM)
L2 Cache
1536 KB
48 MB
SM Count
-
142
Tensor Cores
-
568
RT Cores
-
142
Theoretical Performance
Pixel Rate
42.36 GPixel/s
483.8 GPixel/s
Texture Rate
169.4 GTexel/s
1431.4 GTexel/s
FP32 (float)
4.067 TFLOPS
91.61 TFLOPS
FP64 (double)
1355.5 GFLOPS
1431.4 GFLOPS
FP16 (half)
-
91.61 TFLOPS
Graphics Processor
GPU Name
GK110B
AD102
Architecture
Kepler
Ada Lovelace
Foundry
TSMC
TSMC
Process Size
28 nm
5 nm
Transistors
7.08 billion
76.3 billion
Die Size
561 mm²
609 mm²
GPU Variant
-
AD102-896-A1
Board Design
TDP
245W
350W
Suggested PSU
550 W
750 W
Outputs
No outputs
4x DisplayPort 1.4a
Power Connectors
-
1x 16-pin
Graphics Features
DirectX
12 (11_1)
12 Ultimate (12_2)
OpenGL
4.6
4.6
OpenCL
3.0
3.0
Vulkan
1.2.175
1.4
CUDA
3.5
8.9
Shader Model
6.5 (5.1)
6.9
📊 Benchmark
Blender
-
9,239