NVIDIA B100 vs NVIDIA V100
Side-by-side comparison of specifications, performance, and efficiency.
NVIDIA B100
Architecture
Blackwell
Year
2024
VRAM
192 GB
TDP
700 W
NVIDIA V100
Architecture
Volta
Year
2018
VRAM
32 GB
TDP
250 W
Specifications
NVIDIA B100 NVIDIA V100
Architecture
Blackwell Volta
Launch Year
2024 2018
Form Factor
SXM PCIe
VRAM
192 GB32 GB
Memory Bandwidth
8.0 TB/s900 GB/s
TDP
700 W250 W
Process Node
4nm 12nm
| Spec | NVIDIA B100 | NVIDIA V100 |
|---|---|---|
| Architecture | Blackwell | Volta |
| Launch Year | 2024 | 2018 |
| Form Factor | SXM | PCIe |
| VRAM | 192 GB | 32 GB |
| Memory Bandwidth | 8.0 TB/s | 900 GB/s |
| TDP | 700 W | 250 W |
| Process Node | 4nm | 12nm |
Performance (TFLOPS)
NVIDIA B100 NVIDIA V100
FP64
30.0 TFlops7.0 TFlops
FP32
60.0 TFlops14.0 TFlops
FP16
1750.0 TFlops112.0 TFlops
BF16
1750.0 TFlops—
FP8
3500.0 TFlops—
INT8
3500.0 TFlops—
| Precision | NVIDIA B100 | NVIDIA V100 |
|---|---|---|
| FP64 | 30.0 TFlops | 7.0 TFlops |
| FP32 | 60.0 TFlops | 14.0 TFlops |
| FP16 | 1750.0 TFlops | 112.0 TFlops |
| BF16 | 1750.0 TFlops | — |
| FP8 | 3500.0 TFlops | — |
| INT8 | 3500.0 TFlops | — |
FLOPS by Precision
Efficiency — TDP vs FP32 Performance
Analysis
Most Efficient
NVIDIA
B100
0.09 TFLOPS/W
Get Comparison Updates
New GPUs added weekly. Be the first to see how they compare.