Nous acceptons la cryptomonnaie — les cryptos que vous détenez peuvent ainsi acheter du matériel réel.Payez en crypto, dépensez-la en matériel réel. Livraison assurée offerte partout dans le monde au-delà de 399 $. Cartons neutres, sans marque, expédiés depuis l’Allemagne.Livraison assurée offerte au-delà de 399 $. Votre taux est bloqué pendant 30 minutes dès l’ouverture de la caisse.Taux bloqué 30 minutes. Chaque article est scellé, son numéro de série est vérifié, et il est couvert par une garantie de 24 mois.Scellé, garanti 24 mois.

Articles suivis Se connecter
Full-height graphics card with four DisplayPort outputs and a DVI port

NVIDIA

NVIDIA GP100 graphics card 16 GB HBM2

SKU CH-BXFF-GKSAS MPN GP100

A professional GPU with 16 GB HBM2 memory, 3580 CUDA cores and four DisplayPort outputs for demanding visualisation workloads

  • Memory TypeGDDR5
  • 16 GB HBM2 memory for large datasets
  • 3580 FP32 and 1792 FP64 CUDA cores for compute workloads
  • NVLink support to scale performance across two boards
  • Four DisplayPort outputs for multi-monitor setups
  • PCI Express 3.0 x16 interface for workstation compatibility

Professional compute card

The NVIDIA Quadro GP100 is a full-height graphics card built for professional visualization and simulation workloads. It targets engineers, researchers and developers who need high double-precision throughput and large memory capacity for CAE, rendering and deep-learning tasks.

PCI Express 3.0 x16 interface

The board installs in a standard PCI Express 3.0 x16 slot and draws power from the motherboard and auxiliary connectors. Four DisplayPort outputs drive multiple 3D stereo displays while NVLink allows two GP100 cards to be bridged for shared memory scaling. GPU Direct enables fast data paths to storage and network adapters.

16 GB HBM2 with 4096-bit bus

Sixteen gigabytes of HBM2 memory on a 4096-bit bus deliver the bandwidth needed for large models and datasets. The GPU provides 3580 FP32 and 1792 FP64 CUDA cores for compute-intensive workloads. FP16 performance supports AI training workflows, while the fixed memory capacity sets the upper limit for single-board problem sizes.

Highlights

  • 16 GB HBM2 memory for large datasets
  • 3580 FP32 and 1792 FP64 CUDA cores for compute workloads
  • NVLink support to scale performance across two boards
  • Four DisplayPort outputs for multi-monitor setups
  • PCI Express 3.0 x16 interface for workstation compatibility

Specifications

Memory TypeGDDR5

Questions about this item

How many FP64 CUDA cores limit double-precision throughput on the GP100?

The card provides 1792 FP64 CUDA cores, which set the ceiling for sustained double-precision compute performance.

What is the typical thermal and acoustic behaviour when the Pascal GPU runs all day?

The blower-style cooler exhausts heat out the rear bracket, so case airflow stays cleaner, but the fan ramps up under continuous load and produces audible noise.

Which motherboard slot and power connectors must already be present for the GP100 to operate?

A PCI Express 3.0 x16 slot and two 8-pin PCIe power cables from the power supply are required.

Also in this aisle

People compared these

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