NVIDIA H100 PCIe 80GB
Hopper PCIe 2022 4nm
FP32
51.0
TFLOPS
VRAM
80 GB
TDP
350 W
145.7 TFLOPS/kW
Bandwidth
2.0 TB/s
memory
Performance Metrics
Peak theoretical throughput by precision type
| Precision | Peak TFLOPS | Efficiency |
|---|---|---|
| INT8 | 1513.0 | 4.323 TFLOPS/W |
| FP8 | 1513.0 | 4.323 TFLOPS/W |
| FP16 | 756.5 | 2.161 TFLOPS/W |
| BF16 | 756.5 | 2.161 TFLOPS/W |
| TF32 | 378.0 | 1.080 TFLOPS/W |
| FP32 | 51.0 | 0.146 TFLOPS/W |
| FP64 | 26.0 | 0.074 TFLOPS/W |
FP8 Efficiency
4.323 TFLOPS/W
1513.0 TFLOPS / 350W
FP16 Efficiency
2.161 TFLOPS/W
756.5 TFLOPS / 350W
FP32 Efficiency
0.146 TFLOPS/W
51.0 TFLOPS / 350W
Power Specifications
TDP
350 W
Max Power
402 W
Power Connector
PCIe 16-pin
Cooling
Air
Memory Specifications
Capacity
80 GB
Type
--
Bandwidth
2000 GB/s
Interface
--
Hardware & Design
Form Factor
PCIe
Architecture
Hopper
Process Node
4nm
Launch Year
2022
Variant
PCIe 80GB
Market Segment
Data Center
Documentation & Resources
Common Use Cases
LLM Training LLM Inference Scientific Computing HPC Workloads
The NVIDIA H100 is optimized for high-performance computing tasks with Hopper architecture delivering 51 TFLOPS of compute power.
Where to Rent
Prices shown are approximate. Click through for current pricing.
Similar GPUs
Stay Updated on GPU Releases
Get notified when new GPUs are added or specifications are updated.
Loading verification...
No spam, unsubscribe anytime.