
NVIDIA
RTX PRO 6000 Blackwell
Server Edition
Rack-scale inference, mixed AI and graphics, virtual workstations, and multi-GPU server deployments.
Compare the hardware behind your next deployment. Memory, power, cooling, and platform fit—clearly separated for every model.
Start with the RTX PRO 6000 editions, or select any three cards below.
| Specification | RTX PRO 6000 BlackwellServer Edition | RTX PRO 6000 BlackwellWorkstation Edition | RTX PRO 6000 BlackwellMax-Q Workstation Edition |
|---|---|---|---|
| Architecture | Blackwell | Blackwell | Blackwell |
| GPU memory | 96 GB GDDR7 ECC | 96 GB GDDR7 ECC | 96 GB GDDR7 ECC |
| Memory bandwidth | 1,597 GB/s | 1,792 GB/s | 1,792 GB/s |
| Board power | 400–600 W | 600 W | 300 W |
| Cooling | Passive air; liquid SKU available | Double flow-through | Active |
| Host interface | PCIe 5.0 x16 | PCIe 5.0 x16 | PCIe 5.0 x16 |
| Physical format | Air: dual-slot FHFL · liquid: single-slot FHXL | 5.4″ H × 12″ L · dual slot | 4.4″ H × 10.5″ L · dual slot |
| Deployment | Server | Workstation | Workstation |
| Next step | Quote this edition | Quote this edition | Quote this edition |
Manufacturer specifications reviewed September 11, 2026. Board power is not total system power. Exact part numbers, supported configurations, and availability are confirmed in your quotation.

NVIDIA
Server Edition
Rack-scale inference, mixed AI and graphics, virtual workstations, and multi-GPU server deployments.

NVIDIA
Workstation Edition
Maximum-throughput single-GPU workstations for local AI development, rendering, and demanding visualization.

NVIDIA
Max-Q Workstation Edition
Dense multi-GPU workstations where memory capacity, power, and thermal density matter together.

NVIDIA
48 GB Workstation
Local AI development, rendering, engineering visualization, and memory-intensive desktop applications.

NVIDIA
72 GB Workstation
Local AI development, rendering, engineering visualization, and memory-intensive desktop applications.

NVIDIA
Server Edition
Space-conscious server inference, media processing, analytics, and virtualized GPU services.

NVIDIA
Workstation Edition
Engineering, professional visualization, local inference, and RTX rendering in desktop workstations.

NVIDIA
Workstation
Professional graphics, CAD, video production, and local AI within a single-slot workstation layout.

NVIDIA
SFF Edition
Compact professional workstations for CAD, visualization, and AI-assisted desktop applications.

NVIDIA
Compact workstation
Compact visualization, multi-display professional desktops, and smaller local AI workloads.

NVIDIA
NVL PCIe
Memory-intensive AI inference and HPC on supported PCIe GPU servers.

NVIDIA
SXM / HGX platform
Communication-intensive distributed AI training, large-model inference, and scientific HPC.

NVIDIA
Data-center accelerator
AI inference, rendering, digital twins, and graphics-rich data-center workloads.

NVIDIA
Low-profile server accelerator
Efficient inference, video processing, and GPU services in power-constrained server environments.

NVIDIA
Workstation
Professional rendering, visualization, and AI workflows in existing Ada-based deployments.

NVIDIA
Ampere workstation
Compatible expansion and lifecycle replacement for existing Ampere professional workstations.

NVIDIA
4 GB Workstation
Multi-display business workstations, entry-level CAD and efficient professional desktops.

NVIDIA
8 GB Workstation
Compact professional visualization, light content creation and multi-display workstations.

NVIDIA
6 GB Workstation
Compact CAD and visualization workstations with modest graphics-memory requirements.

NVIDIA
12 GB Workstation
Memory-expanded compact workstations for CAD, visualization and smaller local AI workloads.

NVIDIA
16 GB Workstation
Single-slot professional CAD, rendering and lifecycle replacement for Ampere workstations.

NVIDIA
20 GB Workstation
Engineering visualization, GPU rendering and professional multi-application desktops.

NVIDIA
24 GB Workstation
Professional rendering, virtual production and AI development on existing Ampere platforms.

NVIDIA
24 GB Workstation
High-end Ampere rendering and demanding visualization workflows requiring 24 GB ECC memory.

NVIDIA
16 GB Workstation
Compact CAD, rendering, media work and local AI development within a 70 W power envelope.

NVIDIA
20 GB SFF Workstation
Space-constrained engineering and visualization workstations requiring 20 GB of ECC memory.

NVIDIA
20 GB Workstation
Single-slot CAD, simulation visualization, content creation and local AI workflows.

NVIDIA
24 GB Workstation
Professional rendering, engineering applications and AI development with 24 GB ECC memory.

NVIDIA
32 GB Workstation
Larger 3D scenes, GPU rendering and memory-intensive professional AI development.

NVIDIA
48 GB Workstation
Professional visualization and memory-intensive workloads requiring an exact RTX 5880 SKU.

NVIDIA
84 GB Workstation Edition
Announced · availability pending
Announced rack-mounted workstation GPU for AI, visualization and shared professional compute.
Send your manufacturer SKU, quantity, condition requirement, and delivery schedule.