Data center · PCIe
NVIDIA H100 PCIe 80 GB
The 80 GB PCIe version of the Hopper H100 compute accelerator. It uses HBM2e memory and a Transformer Engine for AI workloads; it is neither the 94 GB H100 NVL nor an H100 SXM module.
Technical specifications
| Architecture | NVIDIA Hopper |
|---|---|
| GPU memory | 80 GB HBM2e |
| Form factor / host interface | PCIe Gen5 |
| CUDA cores | 14592 |
| Tensor cores | 456, fourth generation |
| Compute capability | 9.0 |
| L2 cache | 50 MB |
Applications
- Training and fine-tuning neural networks.
- LLM and other neural-network inference.
- Scientific simulations and GPU-accelerated data analysis.
Compatibility
- Verify the exact server model, slot clearance, power cabling and cooling with the OEM. A PCIe slot alone does not establish compatibility. Passive cooling requires forced airflow from the server.
- H100 is a compute accelerator without display connectors, RT cores or an NVENC encoder. Do not substitute H100 NVL or SXM power and bandwidth figures for this PCIe 80 GB variant.
- Check driver, CUDA/framework and operating-system support for the intended workload. Virtualization and enterprise software may require separate licenses and a supported host configuration.
Manufacturer sources
Specifications checked against the listed NVIDIA sources: .
Request a quote
Each quote is prepared individually for the selected variant, quantity and destination. The RFQ covers condition, configuration, source, warranty, availability and delivery timing.