NVIDIA H100 80GB SXM5
NVIDIA H100 Tensor Core GPU in SXM5 form factor with NVLink for multi-GPU scaling. Designed for HGX server platforms and large-scale AI training clusters. Enterprise volume discounts available - contact sales for custom configurations.
Price Available Upon Request
Custom configuration and pricing
What it's for
The NVIDIA H100 SXM5 delivers the full power of the Hopper architecture with NVLink 4.0 connectivity for unprecedented multi-GPU scaling. Designed for HGX baseboard integration, this flagship configuration enables building supercomputers from 8 to 32,000 GPUs. With 700W TDP and liquid cooling support, H100 SXM unleashes maximum performance for the largest AI training and HPC workloads. Ready to deploy at scale? Our specialists can design custom multi-GPU clusters with complete integration support. Schedule a consultation via the contact form.
Key Features
- ✓80GB HBM3 memory with 3 TB/s bandwidth
- ✓NVLink 4.0 with 900 GB/s bi-directional throughput
- ✓Fourth-generation Tensor Cores with FP8 precision
- ✓1,979 TFLOPS FP8 performance with Transformer Engine
- ✓Scale from 8 to 32,000 GPUs with NVSwitch
- ✓700W TDP for maximum sustained performance
- ✓SXM5 module for direct liquid cooling
- ✓Multi-Instance GPU (MIG) technology for workload isolation
Use Cases
- →Large-scale LLM training (GPT-4 class models)
- →Foundation model development
- →Multi-GPU HPC simulations
- →Scientific research and molecular dynamics
- →Weather forecasting and climate modeling
- →Computational fluid dynamics at scale
Technical Specifications
| Architecture | Hopper |
| GPU Memory | 80 GB HBM3 |
| Memory Bandwidth | 3 TB/s |
| FP64 Performance | 60 TFLOPS |
| FP32 Performance | 120 TFLOPS (TF32) |
| FP16 Performance | ~989 TFLOPS |
| FP8 Performance | ~1,979 TFLOPS |
| INT8 Performance | ~3,958 TOPS |
| CUDA Cores | 16,896 |
| Tensor Cores | 528 (4th Gen) |
| NVLink | 900 GB/s (18 links) |
| Max TDP | 700W |
| Thermal Solution | Passive (liquid cooling required) |
| Form Factor | SXM5 |
| Multi-Instance GPU | Up to 7 instances |
Related Products
NVIDIA H100 80GB PCIe Gen5
NVIDIA H100 Tensor Core GPU in PCIe form factor with 80GB HBM3 memory. Ideal for deploying AI inference and training in standard servers without NVLink clustering requirements. Contact our sales team for volume pricing and current lead time.
From
$29,999
NVIDIA H200 141GB HBM3e SXM5
Hopper GPU with 141GB HBM3e and 4.8TB/s bandwidth — the memory-bandwidth pick for LLM inference. Contact us for current lead time.
From
$39,999
NVIDIA B200 192GB Blackwell
Blackwell architecture with 192GB HBM3e and FP4 precision for next-gen AI. Contact us for availability and pricing.
Request pricing
AMD Instinct MI300X 192GB HBM3 OAM
AMD's flagship AI accelerator with 192GB HBM3 — the largest single-GPU memory in the class — and 5.3TB/s bandwidth. Strong performance per dollar for generative AI and large language models. Quoted per project with ROCm support; lead time confirmed at quote.
From
$29,999
Recommended Reading
The adjacent decisions buyers usually miss
Relevant guides tied to this product's real deployment questions: VRAM, power, cooling, financing, and market timing.
H100 PCIe vs SXM: When Does Form Factor Matter?
Deep dive into H100 PCIe vs SXM differences: NVLink bandwidth, memory throughput, power consumption, and multi-GPU scaling. Decision framework for infrastructure engineers.
GPU Server Power Planning Guide: Circuits, PDUs, and Rack Budgeting
Plan power correctly before your GPU hardware lands. This guide covers circuit sizing, rack density, PDU planning, and realistic power envelopes for H100, H200, MI300X, and B200 systems.
Liquid Cooling for AI GPU Servers: Complete Datacenter Guide
Everything you need to know about liquid cooling for GPU servers: direct-to-chip vs immersion, CDU sizing, retrofit costs ($50K–$150K per row), and which GPUs require it. Essential reading before buying B200 or GB200.