NVIDIA RTX PRO 5000 72GB Blackwell - 900-5G153-2270-000

NVIDIA RTX PRO 5000 72GB Blackwell - 900-5G153-2270-000 - Image 2

NVIDIA RTX PRO 5000 72GB Blackwell – 900-5G153-2270-000

Powered by the Blackwell architecture and 72 GB ECC GDDR7 memory, the NVIDIA RTX PRO 5000 Blackwell Workstation Edition 72 GB is a high-performance professional graphics card designed for AI development, LLM inference, engineering simulations, CAD, 3D rendering, and professional visualization workloads. Featuring PCIe Gen5 connectivity and next-generation Tensor and RT Cores, it accelerates enterprise AI and content creation workflows.

REQUEST FOR QUOTE

Description

NVIDIA RTX PRO 5000 Blackwell Workstation Edition 72 GB combines the next-generation Blackwell architecture developed for professional workstations with high memory capacity and powerful AI acceleration features.

Thanks to its 72 GB ECC GDDR7 memory, it can keep more data directly on the GPU memory while working on large AI models, high-resolution datasets, and complex engineering projects. As a result, it delivers high efficiency in memory-intensive workloads such as LLM inference, generative AI applications, data science, digital twins, CAD/CAM, simulation, and professional content creation.

The next-generation CUDA Cores, Tensor Cores, and RT Cores in the Blackwell architecture offer a significant performance boost in AI inference, real-time ray tracing, physical simulations, and GPU-accelerated professional applications.

Offering high bandwidth thanks to its PCIe Gen5 interface, the RTX PRO 5000 operates with full compatibility in modern workstations, while supporting high-resolution and multi-monitor setups via DisplayPort 2.1b support. Additionally, with the new generation NVENC and NVDEC technologies, it provides faster encoding and decoding performance in video editing, broadcasting, and media workflows.

Fifth Generation Tensor Cores

Fifth-generation Tensor Cores deliver up to three times the performance of the previous generation and provide support for FP4 precision and DLSS 4 Multi-Frame Rendering technology.

4th Generation RT Cores

Fourth-generation RT Cores deliver up to twice the performance of the previous generation, accelerating rendering for M&E content creation, AECO design, and production prototyping.

NVIDIA CUDA Cores

NVIDIA Blackwell, with its latest SM and CUDA core technology, is the most powerful professional RTX GPU ever created.

72 GB GPU Memory

The new and improved GDDR7 memory significantly increases bandwidth and capacity, allowing your applications to run faster and handle larger datasets.

9th Generation NVENC

The ninth-generation NVIDIA NVENC engines significantly accelerate video encoding speed and improve quality for professional video applications. It adds new support for 4:2:2 H.264 and HEVC encoding and improves HEVC and AV1 encoding quality.

6th Generation NVDEC

Sixth-generation NVIDIA NVDEC engines deliver up to twice the H.264 decoding throughput and offer support for 4:2:2 H.264 and HEVC decoding. Professionals can benefit from high-quality video playback, accelerate video data acquisition, and utilize advanced AI-powered video editing features.

PCIe Gen5

PCI Express Gen 5 support delivers twice the bandwidth of PCIe Gen 4, improving data transfer speeds from CPU memory and unlocking faster performance for data-intensive tasks such as AI, data science, and 3D modeling.

Display Port 2.1

With DisplayPort 2.1, professionals can power high-resolution displays up to 8K at 240Hz and 16K at 60Hz, achieving unparalleled visual clarity and performance. Increased bandwidth enables seamless multi-monitor setups ideal for multitasking and collaboration, while HDR and higher color depth support deliver superior color accuracy for demanding tasks such as video editing, 3D design, and live streaming.

Additional information

Card Types	DESKTOP
GPU Architecture	Blackwell
Memory Capacity	72GB
Memory Type	GDDR7
Memory bandwidth	1344 GB/s
DisplayPort Output	Yes
Display connectors	4x DisplayPort 2.1b
System interface	PCIe 5.0 x16
CUDA Cores	14080
Tensor Cores	5th Generation
RT Cores	4th Generation
Max simultaneous displays	2x 7680 x 4320 @ 100 Hz, 4x 3840 x 2160 @ 165 Hz
Video Engines	2x NVDEC (6th Gen), 2x NVENC (9th Gen)
MIG Instance Types	Up to 1x 72 GB, Up to 2x 36 GB
Power consumption	300 W
Power connector	1x PCIe CEM5 16-pin
Thermal solution	Active
Form factor	4.4” H x 10.5” L, dual slot, full height
Graphics APIs	DirectX 12, OpenGL 4.63, Shader Model 6.6, Vulkan 1.3
Compute APIs	CUDA 11.6, CUDA 11.6,, OpenCL 3.0

Jetson

Jetson Thor

AGX Orin

Orin NX

Orin Nano

Servers

DGX

DGX H200

DGX B300

DGX B200

DGX BasePOD

DGX SuperPOD

Custom Clusters

GRAPHICS CARDS

Desktop GPUs

DATA CENTER GPUs

NVIDIA DGX Spark Quad AI Cluster

NVIDIA IGX Thor