We take cryptocurrency — so the coins you hold can buy real hardware.Pay in crypto, spend it on real hardware. Free insured shipping worldwide over $399. Plain, unbranded boxes, sent from Germany.Free insured shipping over $399. Your rate is locked for 30 minutes once checkout opens.Rate locked 30 minutes. Every item sealed, serial-checked and covered by a 24-month warranty.Sealed, 24-month warranty.

Saved items Sign in
Dual-GPU PCIe accelerator card with passive heatsink and gold edge connector

NVIDIA

NVIDIA Tesla K80 24GB GDDR5 CUDA GPU Computing Accelerator Card

SKU CH-GWB6-K968Z

Dual-GPU compute accelerator with 24 GB GDDR5 memory and CUDA support for high-performance workloads

  • Memory TypeGDDR5
  • 24 GB GDDR5 memory supports large datasets
  • CUDA parallel processing accelerates compute workloads
  • PCI Express interface connects to compatible servers
  • Dual-GPU architecture increases throughput for demanding tasks
  • Part number 699-22080-0200-531 identifies the exact model

Dual-GPU compute accelerator

The NVIDIA Tesla K80 is a PCIe add-in card built for GPU computing workloads. It delivers 24 GB of GDDR5 memory and supports CUDA parallel programming for high-throughput tasks.

Workstation and server deployments

This card suits engineers and researchers running CUDA-optimised simulations, rendering or deep-learning training on compatible platforms. It is not intended for gaming systems or consumer workstations without adequate power and cooling provision.

24 GB GDDR5 memory capacity

The 24 GB frame buffer determines the maximum dataset size that fits in local memory. Verify that your application's working set matches this capacity before selecting the K80.

Highlights

  • 24 GB GDDR5 memory supports large datasets
  • CUDA parallel processing accelerates compute workloads
  • PCI Express interface connects to compatible servers
  • Dual-GPU architecture increases throughput for demanding tasks
  • Part number 699-22080-0200-531 identifies the exact model

Specifications

Memory TypeGDDR5

Questions about this item

Does the card fit in a standard PCIe slot?

The card uses a PCIe interface, so it installs in a compatible PCIe slot on the motherboard.

What software framework does the accelerator support for compute workloads?

The card is built for CUDA, allowing it to run CUDA-enabled applications and libraries.

How much memory is available for large datasets?

The board provides 24 GB of GDDR5 memory for data-intensive tasks.

Also in this aisle

People compared these

Same shelf, same checkout — eight coins and a 30-minute rate lock.