A100 80GB SXM4 GPU Accelerator — HBM Data Center Card

An 80GB SXM4 data-center GPU built for AI training and inference at enterprise scale.

From$ 17 000
United States (USD)

Change currency

Prices follow the country you buy from. Pick a country to switch the currency.

Custom order
Made to order to your spec — you choose the size and configuration.

The A100 80GB SXM4 is a data-center GPU accelerator for AI training, inference and high-performance computing. It carries 80GB of HBM in the SXM4 form factor, so it seats into server baseboards that share power and NVLink across multiple cards rather than into a single PCIe slot. That makes it a fit for teams running large models and datasets that spill past the memory of smaller cards.

  • 80GB of high-bandwidth HBM memory for large models and datasets held in-GPU
  • SXM4 socket for dense multi-GPU server nodes
  • Suited to deep-learning training, inference and HPC workloads in enterprise and data-center racks

Units are tested before shipment and ship as new. Lead time is 1–3 days from stock on in-stock configurations; confirm current availability when you order.

Specifications

Memory
80 GB
Memory type
HBM
Form factor
SXM4
Application
AI deep learning, data center, HPC

Frequently asked questions

What does the SXM4 form factor require to run?

SXM4 cards don't drop into a PCIe slot. They seat onto a server baseboard that supplies power and the high-speed interconnect shared across the GPUs, so you need a chassis or HGX-style node built for SXM4 modules rather than a standard tower or workstation.

How much GPU memory does this card have, and why does it matter?

It carries 80GB of high-bandwidth HBM memory. That larger pool keeps big models, larger batch sizes and longer context windows resident on the GPU instead of paging to system RAM, which speeds up both training and inference on large workloads.

What workloads is this A100 accelerator built for?

It targets AI training and inference, deep learning and high-performance computing in enterprise and data-center racks. It's a server-grade accelerator for cooled multi-GPU nodes, not a desktop graphics card.

Is the card new, and is it tested before shipping?

Yes. Each unit is supplied new and tested before dispatch. In-stock configurations typically ship in 1–3 days; confirm current availability and bulk pricing when you order.

Is shipping included in the price?

For in-stock items — anything you can add to the cart — shipping is included in the listed price. For made-to-order items, delivery is either free (marked on the product page) or calculated when you place your order, based on destination and volume.

How long does delivery take?

In-stock items typically arrive within one to three weeks. Made-to-order items take three to six weeks, since we work directly with manufacturers and need time for production and logistics before shipping. Your assigned manager confirms a more precise estimate once your order is placed. Please note we can't be responsible for delays caused by customs clearance in the destination country.

Are import duties and taxes included?

In most cases, yes — we aim to ship Delivered Duty Paid (DDP), which means duties and import taxes are handled so you don't face extra charges on delivery. In rare cases we may ship Delivered Duty Unpaid (DDU), where you would settle import charges directly with the carrier — but only when agreed with you in advance.

Do your products come with a warranty?

Yes. Most of our equipment comes with a one-to-two-year manufacturer's warranty covering factory defects and technical faults under normal use; the specific term is stated on each product page. If a defect appears, we arrange a replacement or refund in line with our Exchange & Return Policy, and our team guides you through the claim step by step.

Request a quote