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 dual-slot graphics card with four DisplayPort connectors and a VirtualLink port

PNY Technologies, Inc.

PNY Technologies, Inc. Quadro RTX 5000 16GB GDDR6 workstation graphics card

SKU CH-SMS1-PRXY0

Workstation GPU with 3072 CUDA and 48 RT cores, 16 GB GDDR6 on a 256-bit bus, and 448 GB/s bandwidth for professional rendering and AI workloads

  • Memory TypeGDDR6
  • 3072 CUDA cores accelerate parallel compute workloads
  • 48 RT cores enable real-time ray-tracing performance
  • 384 Tensor cores deliver 89.2 TFLOPS deep-learning throughput
  • 16 GB GDDR6 memory with 448 GB/s bandwidth supports large datasets
  • Four DP 1.4 outputs plus VirtualLink drive multiple high-resolution displays

Professional GPU for AI and rendering

The PNY Quadro RTX 5000 is a workstation graphics card built on the Turing architecture. It provides 3072 CUDA cores, 384 Tensor cores and 48 RT cores alongside 16 GB of GDDR6 memory on a 256-bit interface delivering 448 GB/s bandwidth. The card connects via PCI Express 3.0 x16 and fits a dual-slot form factor measuring 4.4” H x 10.5” L with an active thermal solution.

Suits deep-learning developers

This card targets professionals who need certified drivers for CAD, simulation and AI workloads on Windows or Linux. It is not for buyers who want more than 16 GB of frame buffer, who require a PCI Express 4.0 or 5.0 link, or who need a single-slot or low-profile design for compact chassis.

Deciding spec: 16 GB GDDR6 frame buffer

The 16 GB GDDR6 memory capacity determines whether the card can hold large models, textures or datasets without paging to system memory. If your workflow fits within this limit the RTX 5000 delivers 11.2 TFLOPS FP32, 22.3 TFLOPS FP16 and 62T RTX-OPS performance within a 265 W power envelope.

Highlights

  • 3072 CUDA cores accelerate parallel compute workloads
  • 48 RT cores enable real-time ray-tracing performance
  • 384 Tensor cores deliver 89.2 TFLOPS deep-learning throughput
  • 16 GB GDDR6 memory with 448 GB/s bandwidth supports large datasets
  • Four DP 1.4 outputs plus VirtualLink drive multiple high-resolution displays

Specifications

Memory TypeGDDR6

Questions about this item

Does the Quadro RTX 5000 come with any cables or adapters in the OEM package?

The OEM package includes the graphics card only. No cables, adapters or brackets are supplied. A compatible power cable from your PSU is required for the 265 W board.

How do I decide between the 16 GB GDDR6 version and a smaller frame buffer?

The 16 GB GDDR6 frame buffer with 448 GB/s bandwidth suits large 3D scenes, 8 K textures and multi-GPU NVLink workloads. Smaller buffers limit dataset size and ray-tracing complexity.

What does the 62T RTX-OPS figure mean for my rendering workflow?

62T RTX-OPS measures combined ray-tracing and AI throughput. In practice it accelerates OptiX rendering, AI denoising and viewport interactivity for complex scenes using the 48 RT cores and 384 Tensor cores.

Also in this aisle

People compared these

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