NVIDIA H800 PCIe 80GB
Overview
China export variant of H100. FP64 capped at 0.8 TFLOPS, NVLink reduced to 400 GB/s.
Performance
Peak theoretical throughput by precision type
| Precision | Dense | 2:4 Sparse |
|---|---|---|
FP64 64-bit floating point | 0.8TFLOPS | Structured sparsity is a tensor-core feature; this vector precision has no sparse form |
FP32 32-bit floating point | 51TFLOPS | Structured sparsity is a tensor-core feature; this vector precision has no sparse form |
TF32 TensorFloat-32 | 378TFLOPS | The part supports 2:4 sparsity, but the vendor publishes no sparse figure for this precision |
BF16 Brain Float 16 | 757TFLOPS | The part supports 2:4 sparsity, but the vendor publishes no sparse figure for this precision |
FP16 16-bit floating point | 757TFLOPS | The part supports 2:4 sparsity, but the vendor publishes no sparse figure for this precision |
FP8 8-bit floating point | 1,513TFLOPS | The part supports 2:4 sparsity, but the vendor publishes no sparse figure for this precision |
FP6 | No verified data available | The part supports 2:4 sparsity, but the vendor publishes no sparse figure for this precision |
FP4 | No verified data available | The part supports 2:4 sparsity, but the vendor publishes no sparse figure for this precision |
INT8 8-bit integer | 1,513TOPS | The part supports 2:4 sparsity, but the vendor publishes no sparse figure for this precision |
Dense peak divided by accelerator TDP. Board power only: excludes host CPUs, networking, cooling and facility overhead. TDP for this part: 350 W.
The H800 PCIe 80GB in the GPU landscape
Peak FP16 TFLOPS (dense) against TDP, single-GPU parts tracked by Flopper
Higher and further left is better: more half-precision throughput for less power.
Specifications
Architecture
Hopper
Form Factor
PCIe
Launch Year
2022
Process Node
4nm
Memory
80 GB HBM3
Bandwidth
2,000 GB/s
TDP
350 W
Max power (Flopper estimate)
~402 W est. Flopper estimate: 350 W TDP x 1.15. The vendor publishes no maximum board power for this part.
Spec Confidence
Official
Full Specifications
| Memory | |
|---|---|
| Memory | 80 GB |
| Memory Type | HBM3 |
| Bandwidth | 2.0 TB/s |
| Interface Width | No verified data available |
| Power & Thermal | |
| TDP | 350 W |
| Max power (Flopper estimate) | ~402 W est. |
| Enterprise Features | |
| Sparsity | Yes |
| General | |
| Form Factor | PCIe |
| Architecture | Hopper |
| Process Node | 4nm |
| Launch Year | 2022 |
Datasheet & Resources
Flopper Datasheet
NVIDIA H800 PCIe 80GB specifications, generated from our database. Printable.
NVIDIA Developer Documentation
Technical specs, programming guides
Data Provenance
Every figure traced to a source
Primary Source
- Document
- —
- Publisher
- NVIDIA
- Published
- No verified data available
Data Quality
- Spec confidence
- Official
- Clock basis
- Boost
- Core precisions with figures
- 7 of 9
- Normalization
- All values in TFLOPS
Compare cloud providers offering on-demand GPU instances for AI training, inference, and HPC workloads.
Browse GPU Cloud ProvidersSimilar GPUs
Frequently Asked Questions
How many TFLOPS does the NVIDIA H800 have?
The NVIDIA H800 delivers 51 TFLOPS FP32, 757 TFLOPS FP16 and 1,513 TFLOPS FP8 at peak.
What is the power consumption of the NVIDIA H800?
The NVIDIA H800 has a TDP (Thermal Design Power) rating of 350 watts.
How much memory does the NVIDIA H800 have?
The NVIDIA H800 is equipped with 80 GB of memory with 2,000 GB/s of memory bandwidth.
What architecture is the NVIDIA H800 based on?
The NVIDIA H800 is based on the Hopper architecture, launched in 2022.
How much does the NVIDIA H800 cost to rent?
Flopper is not tracking any live rental listings for the NVIDIA H800. The provider directory lists the clouds that rent comparable accelerators on demand.
When did the NVIDIA H800 come out?
The NVIDIA H800 launched in 2022 on the Hopper architecture.
Stay Updated on GPU Releases
Get notified when new GPUs are added or specifications are updated.
No spam, unsubscribe anytime.