NVIDIA L4 vs NVIDIA Tesla K20s
We compared two Professional market GPUs: 24GB VRAM L4 and 5GB VRAM Tesla K20s to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
NVIDIA L4 's Advantages
Released 10 years and 1 months late
Boost Clock2040MHz
More VRAM (24GB vs 5GB)
Larger VRAM bandwidth (300.1GB/s vs 208.0GB/s)
4928 additional rendering cores
Lower TDP (72W vs 225W)
Score
Benchmark
FP32 (float)
L4
+955%
30.29 TFLOPS
Tesla K20s
2.87 TFLOPS
Graphics Card
Mar 2023
Release Date
Feb 2013
Tesla Ada(Lxx)
Generation
Tesla
Professional
Type
Professional
PCIe 4.0 x16
Bus Interface
PCIe 2.0 x16
Clock Speeds
795 MHz
Base Clock
-
2040 MHz
Boost Clock
-
1563 MHz
Memory Clock
1300 MHz
Memory
24GB
Memory Size
5GB
GDDR6
Memory Type
GDDR5
192bit
Memory Bus
320bit
300.1GB/s
Bandwidth
208.0GB/s
Render Config
-
-
-
60
SM Count
-
7424
Shading Units
2496
240
TMUs
208
80
ROPs
40
240
Tensor Cores
-
60
RT Cores
-
128 KB (per SM)
L1 Cache
16 KB (per SMX)
48 MB
L2 Cache
1280 KB
-
-
-
Theoretical Performance
163.2 GPixel/s
Pixel Rate
29.90 GPixel/s
489.6 GTexel/s
Texture Rate
119.6 GTexel/s
30.29 TFLOPS
FP16 (half)
-
30.29 TFLOPS
FP32 (float)
2.870 TFLOPS
473.3 GFLOPS
FP64 (double)
956.8 GFLOPS
Board Design
72W
TDP
225W
250 W
Suggested PSU
550 W
No outputs
Outputs
No outputs
None
Power Connectors
-
Graphics Processor
AD104
GPU Name
GK110
AD104-???-A1
GPU Variant
-
Ada Lovelace
Architecture
Kepler
TSMC
Foundry
TSMC
5 nm
Process Size
28 nm
35.8 billion
Transistors
7.08 billion
294 mm²
Die Size
561 mm²
Graphics Features
12 Ultimate (12_2)
DirectX
12 (11_0)
4.6
OpenGL
4.6
3.0
OpenCL
3.0
1.3
Vulkan
1.1
8.9
CUDA
3.5
6.8
Shader Model
5.1
