NVIDIA GRID A100B vs NVIDIA L40 CNX
We compared two Professional market GPUs: 48GB VRAM GRID A100B and 24GB VRAM L40 CNX to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
NVIDIA GRID A100B 's Advantages
More VRAM (48GB vs 24GB)
Larger VRAM bandwidth (1866GB/s vs 864.0GB/s)
NVIDIA L40 CNX 's Advantages
Released 2 years and 5 months late
Boost Clock has increased by 146% (2475MHz vs 1005MHz)
11264 additional rendering cores
Lower TDP (300W vs 400W)
Score
Benchmark
FP32 (float)
GRID A100B
13.89 TFLOPS
L40 CNX
+547%
89.97 TFLOPS
Graphics Card
May 2020
Release Date
Oct 2022
GRID
Generation
Tesla Ada
Professional
Type
Professional
PCIe 4.0 x16
Bus Interface
PCIe 4.0 x16
Clock Speeds
900 MHz
Base Clock
1005 MHz
1005 MHz
Boost Clock
2475 MHz
1215 MHz
Memory Clock
2250 MHz
Memory
48GB
Memory Size
24GB
HBM2e
Memory Type
GDDR6
6144bit
Memory Bus
384bit
1866GB/s
Bandwidth
864.0GB/s
Render Config
-
-
-
108
SM Count
142
6912
Shading Units
18176
432
TMUs
568
192
ROPs
192
432
Tensor Cores
568
-
RT Cores
142
192 KB (per SM)
L1 Cache
128 KB (per SM)
48 MB
L2 Cache
48 MB
-
-
-
Theoretical Performance
193.0 GPixel/s
Pixel Rate
475.2 GPixel/s
434.2 GTexel/s
Texture Rate
1406 GTexel/s
55.57 TFLOPS
FP16 (half)
89.97 TFLOPS
13.89 TFLOPS
FP32 (float)
89.97 TFLOPS
6.947 TFLOPS
FP64 (double)
1406 GFLOPS
Board Design
400W
TDP
300W
800 W
Suggested PSU
700 W
No outputs
Outputs
1x HDMI 2.1
3x DisplayPort 1.4a
None
Power Connectors
1x 16-pin
Graphics Processor
GA100
GPU Name
AD102
-
GPU Variant
AD102-???-A1
Ampere
Architecture
Ada Lovelace
TSMC
Foundry
TSMC
7 nm
Process Size
4 nm
54.2 billion
Transistors
76.3 billion
826 mm²
Die Size
608 mm²
Graphics Features
N/A
DirectX
12 Ultimate (12_2)
N/A
OpenGL
4.6
3.0
OpenCL
3.0
N/A
Vulkan
1.3
8.0
CUDA
8.9
N/A
Shader Model
6.6
