NVIDIA L40 vs NVIDIA L4
We compared two Professional market GPUs: 48GB VRAM L40 and 24GB VRAM L4 to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
NVIDIA L40 's Advantages
Boost Clock has increased by 22% (2490MHz vs 2040MHz)
More VRAM (48GB vs 24GB)
Larger VRAM bandwidth (864.0GB/s vs 300.1GB/s)
10752 additional rendering cores
NVIDIA L4 's Advantages
Lower TDP (72W vs 300W)
Score
Benchmark
FP32 (float)
L40
+198%
90.52 TFLOPS
L4
30.29 TFLOPS
Graphics Card
Oct 2022
Release Date
Mar 2023
Tesla Ada
Generation
Tesla Ada(Lxx)
Professional
Type
Professional
PCIe 4.0 x16
Bus Interface
PCIe 4.0 x16
Clock Speeds
735 MHz
Base Clock
795 MHz
2490 MHz
Boost Clock
2040 MHz
2250 MHz
Memory Clock
1563 MHz
Memory
48GB
Memory Size
24GB
GDDR6
Memory Type
GDDR6
384bit
Memory Bus
192bit
864.0GB/s
Bandwidth
300.1GB/s
Render Config
-
-
-
142
SM Count
60
18176
Shading Units
7424
568
TMUs
240
192
ROPs
80
568
Tensor Cores
240
142
RT Cores
60
128 KB (per SM)
L1 Cache
128 KB (per SM)
96 MB
L2 Cache
48 MB
-
-
-
Theoretical Performance
478.1 GPixel/s
Pixel Rate
163.2 GPixel/s
1414 GTexel/s
Texture Rate
489.6 GTexel/s
90.52 TFLOPS
FP16 (half)
30.29 TFLOPS
90.52 TFLOPS
FP32 (float)
30.29 TFLOPS
1414 GFLOPS
FP64 (double)
473.3 GFLOPS
Board Design
300W
TDP
72W
700 W
Suggested PSU
250 W
4x DisplayPort 1.4a
Outputs
No outputs
1x 16-pin
Power Connectors
None
Graphics Processor
AD102
GPU Name
AD104
AD102-895-A1
GPU Variant
AD104-???-A1
Ada Lovelace
Architecture
Ada Lovelace
TSMC
Foundry
TSMC
4 nm
Process Size
5 nm
76.3 billion
Transistors
35.8 billion
608 mm²
Die Size
294 mm²
Graphics Features
12 Ultimate (12_2)
DirectX
12 Ultimate (12_2)
4.6
OpenGL
4.6
3.0
OpenCL
3.0
1.3
Vulkan
1.3
8.9
CUDA
8.9
6.6
Shader Model
6.8
