NVIDIA A100 PCIe vs NVIDIA A100 SXM4 80 GB
AI GPUAI GPU
We compared two GPUs: 40GB VRAM A100 PCIe and 80GB VRAM A100 SXM4 80 GB to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
NVIDIA A100 PCIe 's Advantages
Lower TDP (250W vs 400W)
NVIDIA A100 SXM4 80 GB 's Advantages
More VRAM (80GB vs 40GB)
Larger VRAM bandwidth (2039GB/s vs 1555GB/s)
Score
Benchmark
FP32 (float)
A100 PCIe
19.49 TFLOPS
A100 SXM4 80 GB
19.49 TFLOPS
Graphics Card
Jun 2020
Release Date
Nov 2020
Tesla
Generation
Tesla Ampere
AI GPU
Type
AI GPU
PCIe 4.0 x16
Bus Interface
PCIe 4.0 x16
Clock Speeds
765 MHz
Base Clock
1275 MHz
1410 MHz
Boost Clock
1410 MHz
1215 MHz
Memory Clock
1593 MHz
Memory
40GB
Memory Size
80GB
HBM2e
Memory Type
HBM2e
5120bit
Memory Bus
5120bit
1555GB/s
Bandwidth
2039GB/s
Render Config
-
-
-
108
SM Count
108
6912
Shading Units
6912
432
TMUs
432
160
ROPs
160
432
Tensor Cores
432
-
-
-
192 KB (per SM)
L1 Cache
192 KB (per SM)
40 MB
L2 Cache
40 MB
-
-
-
Theoretical Performance
225.6 GPixel/s
Pixel Rate
225.6 GPixel/s
609.1 GTexel/s
Texture Rate
609.1 GTexel/s
77.97 TFLOPS
FP16 (half)
77.97 TFLOPS
19.49 TFLOPS
FP32 (float)
19.49 TFLOPS
9.746 TFLOPS
FP64 (double)
9.746 TFLOPS
Board Design
250W
TDP
400W
600 W
Suggested PSU
800 W
No outputs
Outputs
No outputs
8-pin EPS
Power Connectors
None
Graphics Processor
GA100
GPU Name
GA100
-
-
-
Ampere
Architecture
Ampere
TSMC
Foundry
TSMC
7 nm
Process Size
7 nm
54.2 billion
Transistors
54.2 billion
826 mm²
Die Size
826 mm²
Graphics Features
N/A
DirectX
N/A
N/A
OpenGL
N/A
3.0
OpenCL
3.0
N/A
Vulkan
N/A
8.0
CUDA
8.0
N/A
Shader Model
N/A
