NVIDIA A40 vs NVIDIA V100
Side-by-side comparison of specifications, performance, and efficiency.
NVIDIA A40
OfficialArchitecture
Ampere
Year
2021
VRAM
48 GB
TDP
300 W
NVIDIA V100
OfficialArchitecture
Volta
Year
2017
VRAM
16 GB
TDP
250 W
Specifications
NVIDIA A40 NVIDIA V100
Architecture
Ampere Volta
Launch Year
2021 2017
Form Factor
PCIe PCIe
VRAM
48 GB16 GB
Memory Bandwidth
696 GB/s900 GB/s
TDP
300 W250 W
Process Node
— 12nm
| Spec | NVIDIA A40 | NVIDIA V100 |
|---|---|---|
| Architecture | Ampere | Volta |
| Launch Year | 2021 | 2017 |
| Form Factor | PCIe | PCIe |
| VRAM | 48 GB | 16 GB |
| Memory Bandwidth | 696 GB/s | 900 GB/s |
| TDP | 300 W | 250 W |
| Process Node | — | 12nm |
Performance (TFLOPS)
NVIDIA A40 NVIDIA V100
FP64
—7.0 TFlops
FP32
37.4 TFlops14.0 TFlops
FP16
149.7 TFlops112.0 TFlops
BF16
149.7 TFlops—
INT8
299.3 TFlops—
| Precision | NVIDIA A40 | NVIDIA V100 |
|---|---|---|
| FP64 | — | 7.0 TFlops |
| FP32 | 37.4 TFlops | 14.0 TFlops |
| FP16 | 149.7 TFlops | 112.0 TFlops |
| BF16 | 149.7 TFlops | — |
| INT8 | 299.3 TFlops | — |
FLOPS by Precision
Efficiency — TDP vs FP32 Performance
Analysis
Most Efficient
NVIDIA
A40
0.12 TFLOPS/W
Get Comparison Updates
New GPUs added weekly. Be the first to see how they compare.