NVIDIA RTX PRO 5000 72GB Blackwell GPU now generally available

Top-notch options for AI at the desktops of developers, engineers and designers are expanding.

The NVIDIA RTX PRO 5000 72GB Blackwell GPU is now generally available, bringing robust agentic and generative AI capabilities powered by the NVIDIA Blackwell architecture to more desktops and professionals across the world.

The new GPU configuration offers AI developers, data scientists and creative professionals the hardware for modern, memory-hungry workflows — and arrives at a time when demand for NVIDIA Blackwell-class compute is higher than ever.

With the flexibility to choose between this 72GB variant and the existing NVIDIA RTX PRO 5000 48GB model, AI developers can right-size their systems for a wider range of budgets and project requirements.


The NVIDIA RTX PRO 5000 Blackwell GPU.

Fueling the Next Generation of AI Development​

As generative AI evolves into complex, multimodal agentic AI, more demand is placed on the hardware required to develop and deploy these technologies.

One defining challenge of AI development is memory capacity. Running cutting-edge AI workflows — especially those involving large language models (LLMs) and AI agents — places significant stress on GPU memory, particularly as models, context windows and multimodal pipelines grow in size and complexity.

Agentic AI systems involve chains of tools, retrieval-augmented generation (RAG), and multimodal understanding. These systems often need to keep multiple AI models, data sources and code formats active simultaneously within the GPU’s memory.

Built on NVIDIA Blackwell — which delivers high throughput for AI, neural rendering and simulation with multi-workload scheduling and other architectural innovations — the RTX PRO 5000 72GB helps solve this bottleneck, offering 2,142 TOPS of AI performance.

Plus, with 72GB of ultrafast GDDR7 memory — a 50% increase over the 48GB model — developers can train, fine-tune and prototype larger models locally.

This enables users to maintain data privacy, low latency and cost efficiency, allowing teams to serve models directly from their workstations rather than relying on data-center-scale infrastructure for every AI task.

Performance Enhancements​

For local AI development, raw compute is only half the battle — memory capacity determines what users can run, and throughput determines how fast it runs.

In industry-standard benchmarks for generative AI, the RTX PRO 5000 72GB offers 3.5x the performance of prior-generation NVIDIA hardware for image generation, and 2x the performance of prior-generation hardware for text generation.



In creative workflows, time saved in rendering is time gained for iteration. Across path-tracing engines like Arnold, Chaos V-Ray and Blender, as well as real-time GPU renderers like D5 Render and Redshift, the RTX PRO 5000 72GB slashes render times by up to 4.7x.

And for computer-aided engineering and product design, the RTX PRO 5000 72GB offers more than 2x graphics performance.

InfinitForm​

InfinitForm, a provider of generative AI software for engineering design and a member of the NVIDIA Inception program for startups, is an early adopter of the RTX PRO 5000 72GB Blackwell GPU.

The company is using the GPU to optimize its software with enhanced performance and speed, enabling advanced simulations to streamline processes for computer-aided design and manufacturing.

“InfinitForm is thrilled to evaluate its CUDA-accelerated generative AI design optimization software on NVIDIA RTX PRO 5000 72GB to help customers like Yamaha Motor and NASA accelerate innovation and optimize products for performance and manufacturability,” said Michael Bogomolny, founder and CEO of InfinitForm.

Versatile Media Accelerates Virtual Production Workflows With RTX PRO 5000​

For creative professionals, such as those at Versatile Media — a global media company specializing in virtual production — the RTX PRO 5000 can significantly enhance real-time rendering performance for large-scale scenes and complex assets, delivering a meaningful leap in efficiency.

As 3D pipelines now often integrate AI denoisers, generative tools and real-time physics, the new GPU’s 72GB memory capacity allows for the manipulation of massive 3D scenes and asset libraries without slowing down the creative flow.

An early adopter of the new GPU, Versatile Media plans to use it to design a series of complex, high-resolution real-time rendering scenarios that take advantage of the GPU’s memory capacity.

“For film-grade virtual production, memory capacity directly translates into creative freedom,” said Eddy Shen, general manager of the production center at Versatile Media. “With 72GB of GPU memory, the RTX PRO 5000 enables us to iterate with higher-resolution scenes and more complex lighting in real time without compromising performance.”

Availability and Ecosystem Support​

The NVIDIA RTX PRO 5000 72GB Blackwell GPU is now generally available from partners including Ingram Micro, Leadtek, Unisplendour and xFusion, providing manufacturers and systems integrators with a powerful new option to anchor AI-ready workstations. Broader availability through global system builders will start early next year.

As industries race to integrate AI into every facet of operation — from generative design to coding copilots — RTX PRO 5000 72GB is equipped to meet the moment.

Learn more about the NVIDIA RTX PRO 5000 GPU family and join NVIDIA at SIGGRAPH Asia to discover how graphics and simulation innovations come together to drive creativity, industrial digitalization, robotics, and physical and spatial AI.