PNY NVIDIA H100 80GB SXM5 GPU Compute Module

PNY NVIDIA H100 80GB SXM5 GPU Compute Module

Brand: PNY | Category: GPUs

SKU: TCSH10080GB-SXM5-LR | Part #: TCSH10080GB-SXM5-LR | MPN: TCSH10080GB-SXM5-LR

Contact for Pricing — Request a Quote

Request a Quote Contact Us

About the PNY NVIDIA H100 80GB SXM5 GPU Compute Module

PNY NVIDIA H100 80GB SXM5 GPU Compute Module is the flagship AI training accelerator for hyperscale data centers and enterprise AI infrastructure. The SXM5 form factor enables direct NVLink connectivity and optimal thermal design for the most demanding transformer model training, distributed deep learning, and next-generation AI applications.

Key Features:

  • 80GB HBM3 memory with world-leading training throughput
  • SXM5 socket for NVLink GPU-to-GPU communication at 900GB/s
  • Tensor float 32 and transformer engine optimizations for LLM training
  • Multi-instance GPU (MIG) support for workload consolidation

Perfect for:

  • Large-scale AI model training and foundation model development
  • Hyperscale cloud infrastructure and AI service providers
  • Enterprise deep learning platforms and research laboratories
  • Multi-GPU cluster deployments with NVLink coherence

Why source from Omnixon Global:

  • Authorized channel distributor — 100% authentic, channel-direct
  • Competitive pricing with volume discounts for enterprise buyers
  • Dubai hub with rapid fulfillment to UAE, Saudi Arabia, Qatar, Kuwait, Oman, Bahrain and worldwide
  • Request a quote via the RFQ form or WhatsApp for lead time and configuration options

Technical Specifications

BrandPNY
CategoryGPUs
SKUTCSH10080GB-SXM5-LR
Part NumberTCSH10080GB-SXM5-LR
ConditionNew
Form FactorSXM5 Module
GPU Memory80GB HBM3
Memory InterfaceHBM3
Peak FP32 Performance1979 TFLOPS
Peak TF32 Performance1979 TFLOPS
Peak FP64 Performance989.5 TFLOPS
Memory Bandwidth3.35 TB/s
L2 Cache50 MB
Max Power Consumption700W
NVLink12x NVLink 4.0 (900 GB/s aggregate)
GPU ArchitectureNVIDIA Hopper
Supported FrameworksCUDA 12.x, cuDNN, TensorRT, Triton