Powering the next era of AI.
Overview
The NVIDIA RTX PRO™ 5500 Blackwell Workstation Edition accelerates agentic AI, physical simulation, and graphics workloads to teams that need workstation power at scale. Built on the NVIDIA Blackwell architecture and equipped with 84 GB of ultra-fast GDDR7 memory, it’s designed for rack-mounted workstation deployment with a choice of air- or liquid-cooled thermal solutions, letting teams centralize, share, and scale workstation GPU power across the enterprise.
Features
Workloads
Specifications*
| GPU Architecture | NVIDIA Blackwell |
|---|---|
| Tensor Cores | 5th Generation |
| Ray Tracing Cores | 4th Generation |
| NVIDIA Encoder (NVENC) | 3x 9th Generation |
| NVIDIA Decoder (NVDEC) | 3x 6th Generation |
| GPU Memory | 84 GB GDDR7 with error-correcting code (ECC) |
| Memory Bandwidth | 1398 GB/sec |
| Display Ports | Up to 4x DisplayPort 2.1b |
| Max Power Consumption | Up to 600 W |
| Graphics Bus | PCIe Gen 5 x16 |
| Thermal | Active, air-cooled** |
*Preliminary product specifications, subject to change.
**Liquid-cooled solution is available in the RXM form factor.
The NVIDIA RTX PRO 5500 Blackwell Workstation Edition is a professional desktop GPU built on the NVIDIA Blackwell architecture with 84 GB of GDDR7 memory. It accelerates agentic AI, physical AI, simulation, and graphics workloads, and is designed for rack-mounted workstation deployment across the enterprise.
The RTX PRO 5500 Blackwell targets multi-workload acceleration: agentic AI, generative AI and LLM inference, physical AI and simulation, AI-powered rendering, scientific computing and data analytics, and video production. A single card serves teams needing workstation power for both AI and graphics.
The RTX PRO 5500 Blackwell is built for rack-mounted workstation deployment, with a choice of air- or liquid-cooled thermal solutions. IT teams can centralize GPUs into managed racks, share them across users, and scale capacity as demand grows—delivering workstation performance to professionals without individual desk-side hardware.
The RTX PRO 5500 Blackwell features 84 GB of GDDR7 memory with error-correcting code (ECC). The large GPU memory enables professionals to work with larger, more complex datasets—running large AI models, simulations, and detailed 3D scenes without splitting work across systems.
MIG partitions a single RTX PRO 5500 into fully isolated instances—one instance with up to 84 GB, or two instances with up to 42 GB each. Every instance has dedicated memory, cache, and compute cores plus guaranteed quality of service. In rack-mounted deployments, that means more users get reliable accelerated computing from the same hardware.
The RTX PRO 5500 Blackwell offers a choice of air-cooled or liquid-cooled thermal solutions for rack-mounted workstation deployment.
Yes. The RTX PRO 5500 Blackwell powers LLM inference, AI agents, generative AI, and computer vision from the desktop. 84 GB of GDDR7 memory holds larger models, longer context windows, and multiple models at once—so agentic pipelines run without offloading. Fifth-gen Tensor Cores add FP4 precision and deliver up to 3X the performance of the previous generation for multimodal agentic and generative AI. Learn more.
Fifth-gen Tensor Cores, fourth-gen RT Cores, and 84 GB of GDDR7 memory deliver the throughput physical AI demands—running high-fidelity simulation, generating synthetic training data, and developing and testing robotics workflows before anything is built in the real world.
The RTX PRO 5500 Blackwell is designed for professionals in media and entertainment, architecture and engineering (AEC), manufacturing, engineering, higher education, financial services, geoscience, and healthcare—wwherever AI, simulation, rendering, scientific computing, and data analytics demand workstation-class performance at scale.
The RTX PRO 5500 Blackwell is coming soon; you can sign up to be notified when it becomes available. To evaluate fit, contact an NVIDIA product expert or sign up for the latest RTX PRO news.
