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
Silver single-slot graphics card with four DisplayPort outputs and a blower-style cooler

NVIDIA

NVIDIA RTX 4000 Ada 20GB graphics card

SKU CH-HPQT-EPSTE

Single-slot workstation GPU with 20 GB GDDR6 on a 160-bit bus, 6144 CUDA cores, PCIe 4.0 x16, four DisplayPort 1.4a outputs and AV1 encode/decode support

  • InterfacePCI Express 4.0 x16
  • Chipset ManufacturerNVIDIA
  • GPU SeriesRTX 4000 Ada
  • CUDA Cores6144
  • Memory Size20 GB
  • Memory Interface160-Bit
  • 20 GB GDDR6 memory handles large datasets and complex scenes
  • 6144 CUDA cores accelerate rendering, simulation and AI workloads
  • PCI Express 4.0 x16 interface provides high-bandwidth data transfer
  • Four DisplayPort 1.4a outputs drive multiple high-resolution displays
  • AV1 encode and decode support speeds up modern video workflows

Professional GPU for visual computing

The NVIDIA RTX 4000 Ada 20GB is a workstation graphics card built on the Ada architecture for professional visual computing. It delivers 6144 CUDA cores and 20 GB of GDDR6 memory across a 160-bit interface connected through PCI Express 4.0 x16. The single-slot design suits dense workstations and rack deployments where space is limited. Windows and Linux operating systems are supported for broad software compatibility.

Thermal and power profile

A single-slot cooler manages heat within the constraints of a compact form factor while the card draws power appropriate for professional workloads. Fan speed ramps under sustained compute or render tasks and settles to a low acoustic level during typical desktop or viewport use. The 160-bit memory bus and Ada efficiency help keep power draw proportional to the work performed. Driver controls let users tune thermal targets for quieter or cooler operation as needed.

Build quality and connectivity

Four DisplayPort 1.4a outputs drive multiple high-resolution displays with audio and HDCP 2.2 support for protected content. A dedicated stereo connector enables 3D stereo workflows in compatible applications. GPUDirect for Video and RDMA allow direct data paths between the GPU and capture or storage devices, reducing latency in broadcast and scientific pipelines. Quadro Sync II compatibility adds frame-lock across cards for multi-display installations.

Highlights

  • 20 GB GDDR6 memory handles large datasets and complex scenes
  • 6144 CUDA cores accelerate rendering, simulation and AI workloads
  • PCI Express 4.0 x16 interface provides high-bandwidth data transfer
  • Four DisplayPort 1.4a outputs drive multiple high-resolution displays
  • AV1 encode and decode support speeds up modern video workflows

Specifications

BrandNVIDIA
SeriesNVIDIA RTX 4000 Ada 20GB
ModelVCNRTX4000ADA-PB 20GB
InterfacePCI Express 4.0 x16
Chipset ManufacturerNVIDIA
GPU SeriesRTX 4000 Ada
CUDA Cores6144
Memory Size20 GB
Memory Interface160-Bit
DirectXDirectX 12
OpenGLOpenGL 4.6
DisplayPort4
Operating Systems SupportedWindows, Linux
Slot WidthSingle Slot

Questions about this item

How do I install the RTX 4000 Ada 20GB in a PCI Express 4.0 x16 slot without bending the single-slot bracket?

Align the card's PCIe 4.0 x16 edge connector with the slot, press straight down until the retention clip clicks, then secure the single-slot bracket with one screw; the common mistake is angling the card during insertion, which can bend the thin bracket or damage the connector.

What maintenance will the 20 GB GDDR6 memory and cooling assembly need over time?

Dust the heatsink fins and fan blades with compressed air every few months; the 20 GB GDDR6 modules are soldered and require no user service, while the fan bearing is sealed and not user-replaceable.

At sustained load, does the 160-bit memory interface or the 6144 CUDA cores hit their limit first?

The 160-bit memory interface typically saturates before the 6144 CUDA cores reach peak utilisation, making memory bandwidth the first bottleneck in prolonged compute or render workloads.

Also in this aisle

People compared these

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