NVIDIA H200 NVL 141GB
Hopper PCIe 2024 4nm
FP32
60.0
TFLOPS
VRAM
141 GB
HBM3e
TDP
600 W
100.0 TFLOPS/kW
Bandwidth
4.8 TB/s
memory
Performance Metrics
Peak theoretical throughput by precision type
| Precision | Description | Bits | Peak TFLOPS | Efficiency |
|---|---|---|---|---|
| INT8 | 8-bit integer | 8 | 1670.5 | 2.784 TFLOPS/W |
| FP8 | 8-bit floating point | 8 | 1670.5 | 2.784 TFLOPS/W |
| FP16 | 16-bit floating point | 16 | 835.5 | 1.393 TFLOPS/W |
| BF16 | Brain Float 16 | 16 | 835.5 | 1.393 TFLOPS/W |
| TF32 | TensorFloat-32 | 32 | 417.5 | 0.696 TFLOPS/W |
| FP32 | 32-bit floating point | 32 | 60.0 | 0.100 TFLOPS/W |
| FP64 | 64-bit floating point | 64 | 30.0 | 0.050 TFLOPS/W |
FP8 Efficiency
2.784 TFLOPS/W
1670.5 TFLOPS / 600W
FP16 Efficiency
1.393 TFLOPS/W
835.5 TFLOPS / 600W
FP32 Efficiency
0.100 TFLOPS/W
60.0 TFLOPS / 600W
Power Specifications
TDP
600 W
Max Power
690 W
Power Connector
PCIe 16-pin
Cooling
Air
Memory Specifications
Capacity
141 GB
Type
HBM3e
Bandwidth
4800 GB/s
Interface
--
Hardware & Design
Form Factor
PCIe
Architecture
Hopper
Process Node
4nm
Launch Year
2024
Variant
NVL 141GB
Market Segment
Data Center
Common Use Cases
LLM Training LLM Inference Scientific Computing HPC Workloads
The NVIDIA H200 is optimized for high-performance computing tasks with Hopper architecture delivering 60 TFLOPS of compute power.
Where to Rent
Compare cloud providers offering on-demand GPU instances for AI training, inference, and HPC workloads.
Browse GPU Cloud ProvidersSimilar GPUs
Stay Updated on GPU Releases
Get notified when new GPUs are added or specifications are updated.
Loading verification...
No spam, unsubscribe anytime.