Limited Time Offer: Save $350 on your Ai Cluster → Shop Now

--

Days

--

Hours

--

Mins

--

Secs

Sale Ended

PNY NVIDIA RTX™ PRO 4000 Blackwell - 24GB GDDR7
PNY NVIDIA RTX™ PRO 4000 Blackwell - 24GB GDDR7

Vendor:PNY

PNY NVIDIA RTX™ PRO 4000 Blackwell - 24GB GDDR7

$3,499.00

The PNY NVIDIA RTX PRO 4000 Blackwell is a single-slot professional graphics card for local AI, 3D design, engineering, rendering, video production, and visualization. Built on NVIDIA Blackwell architecture, it combines 8,960 CUDA cores, 24GB of ECC GDDR7 memory, fifth-generation Tensor Cores, and fourth-generation RT Cores in a 140W full-height card. PCIe Gen5, four DisplayPort 2.1b connections, and next-generation encode and decode engines make it a strong fit for compact workstations and high-density professional deployments.

Out of stock
Product datasheet

Pairs well with

30-Day AISuperStore Guarantee 

TECHNOLOGY AT A GLANCE

Explore the capabilities

24GB ECC GDDR7 memory

High-speed error-correcting memory supports larger datasets, scenes, models, and multi-application workflows.

8,960 CUDA cores

NVIDIA Blackwell CUDA cores accelerate compatible professional rendering, simulation, visualization, and compute tasks.

5th Gen Tensor Cores

Support FP4 precision and accelerate compatible generative AI, local model inference, and neural graphics workflows.

Single-slot 140W design

A full-height, active-cooled single-slot card fits compatible compact workstations and high-density systems.

Professional Blackwell performance in a single-slot GPU

RTX PRO 4000 Blackwell brings 24GB of ECC GDDR7 memory, accelerated AI and ray tracing hardware, and professional display connectivity to compact workstation environments.

TECHNICAL DETAILS

Built into the system

NVIDIA Blackwell GPU architecture

The RTX PRO 4000 is built on NVIDIA Blackwell architecture, bringing advanced AI processing, neural rendering, ray tracing, and professional graphics capabilities to a single-slot workstation GPU.

24GB ECC GDDR7 memory

24GB of GDDR7 memory with ECC helps support larger professional datasets, detailed 3D scenes, AI models, high-resolution assets, and multi-application workflows. The card provides up to 672GB/s of memory bandwidth.

AI and CUDA acceleration

8,960 CUDA cores and fifth-generation Tensor Cores support compatible AI inference, generative AI, local model fine-tuning, simulation, rendering, and compute workflows. Tensor Cores support FP4 precision for supported workloads.

Real-time ray tracing and neural graphics

Fourth-generation RT Cores accelerate compatible ray-traced rendering and visualization. NVIDIA Blackwell neural-shader capabilities support new AI-assisted graphics workflows in compatible applications.

Professional display and video workflows

Four DisplayPort 2.1b connectors support compatible multi-display configurations, including up to four 4K displays at 165Hz or two 8K displays at 100Hz. Two ninth-generation NVENC encoders and two sixth-generation NVDEC decoders support compatible high-resolution video workflows, including 4:2:2 encoding and decoding.

Compact workstation integration

The card uses a PCIe 5.0 x16 interface, measures 4.4 × 9.5 inches, and uses a full-height, single-slot active-cooled design. Total board power is 140W through one PCIe CEM5 16-pin power connector.