NVIDIA B100 vs NVIDIA H100
Side-by-side comparison of specifications, performance, and efficiency.
NVIDIA B100
OfficialArchitecture
Blackwell
Year
2024
VRAM
192 GB
TDP
700 W
NVIDIA H100
OfficialArchitecture
Hopper
Year
2022
VRAM
80 GB
TDP
350 W
Specifications
NVIDIA B100 NVIDIA H100
Architecture
Blackwell Hopper
Launch Year
2024 2022
Form Factor
SXM PCIe
VRAM
192 GB80 GB
Memory Bandwidth
8.0 TB/s2.0 TB/s
TDP
700 W350 W
Process Node
4nm 4nm
| Spec | NVIDIA B100 | NVIDIA H100 |
|---|---|---|
| Architecture | Blackwell | Hopper |
| Launch Year | 2024 | 2022 |
| Form Factor | SXM | PCIe |
| VRAM | 192 GB | 80 GB |
| Memory Bandwidth | 8.0 TB/s | 2.0 TB/s |
| TDP | 700 W | 350 W |
| Process Node | 4nm | 4nm |
Performance (TFLOPS)
NVIDIA B100 NVIDIA H100
FP64
30.0 TFlops26.0 TFlops
FP32
60.0 TFlops51.0 TFlops
FP16
1750.0 TFlops756.5 TFlops
BF16
1750.0 TFlops756.5 TFlops
FP8
3500.0 TFlops1513.0 TFlops
INT8
3500.0 TFlops1513.0 TFlops
| Precision | NVIDIA B100 | NVIDIA H100 |
|---|---|---|
| FP64 | 30.0 TFlops | 26.0 TFlops |
| FP32 | 60.0 TFlops | 51.0 TFlops |
| FP16 | 1750.0 TFlops | 756.5 TFlops |
| BF16 | 1750.0 TFlops | 756.5 TFlops |
| FP8 | 3500.0 TFlops | 1513.0 TFlops |
| INT8 | 3500.0 TFlops | 1513.0 TFlops |
FLOPS by Precision
Efficiency — TDP vs FP32 Performance
Analysis
Most Efficient
NVIDIA
H100
0.15 TFLOPS/W
Get Comparison Updates
New GPUs added weekly. Be the first to see how they compare.