



Leadtek NVIDIA RTX PRO 5000 Blackwell 48GB GDDR7 ECC GPU
The Leadtek NVIDIA RTX PRO 5000 Blackwell GPU is built on the NVIDIA Blackwell architecture, delivering exceptional performance for professional design, data science, simulation, and AI workflows. It features 14,080 CUDA cores, 440 Tensor Cores, and 110 RT Cores, providing up to 2,142 AI TOPS for accelerated computing.
Equipped with 48GB GDDR7 memory with ECC support on a 384-bit interface, this GPU offers a memory bandwidth of 1,344 GB/s. It supports PCI Express 5.0 x16 interface and includes four DisplayPort 2.1 connectors for high-resolution displays.
The card has a full-height full-length dual-slot form factor measuring 4.4 inches by 10.5 inches, with a blower active fan thermal solution. Power is supplied via a single PCIe CEM5 16-pin connector, with a maximum power consumption of 300W. It also supports Multi-Instance GPU configurations and is compatible with NVIDIA RTX PRO Sync for frame lock.
| Architecture | NVIDIA Blackwell |
| CUDA Parallel Processing Cores | 14,080 |
| NVIDIA Tensor Cores | 440 |
| NVIDIA RT Cores | 110 |
| AI Performance | 2,142 AI TOPS |
| GPU Memory | 48 GB GDDR7 with ECC |
| Memory Interface | 384-bit |
| Memory Bandwidth | 1,344 GB/s |
| Max Power Consumption | 300W |
| Multi-Instance GPU | Up to 2x 24GB Up to 1x 48GB |
| Graphics Bus | PCI Express 5.0 x16 |
| Display Connectors | DP 2.1 (4) |
| Form Factor | 4.4" H x 10.5" L, FHFL Dual Slot |
| Product Weight | 1.219 kg |
| Thermal Solution | Blower Active Fan |
| Power Connector | 1x PCIe CEM5 16-pin |
| NVENC | NVDEC | JPEG | 3x | 3x | 1x |
Product Information
Product Information
Shipping & Returns
Shipping & Returns
Description
The Leadtek NVIDIA RTX PRO 5000 Blackwell GPU is built on the NVIDIA Blackwell architecture, delivering exceptional performance for professional design, data science, simulation, and AI workflows. It features 14,080 CUDA cores, 440 Tensor Cores, and 110 RT Cores, providing up to 2,142 AI TOPS for accelerated computing.
Equipped with 48GB GDDR7 memory with ECC support on a 384-bit interface, this GPU offers a memory bandwidth of 1,344 GB/s. It supports PCI Express 5.0 x16 interface and includes four DisplayPort 2.1 connectors for high-resolution displays.
The card has a full-height full-length dual-slot form factor measuring 4.4 inches by 10.5 inches, with a blower active fan thermal solution. Power is supplied via a single PCIe CEM5 16-pin connector, with a maximum power consumption of 300W. It also supports Multi-Instance GPU configurations and is compatible with NVIDIA RTX PRO Sync for frame lock.
| Architecture | NVIDIA Blackwell |
| CUDA Parallel Processing Cores | 14,080 |
| NVIDIA Tensor Cores | 440 |
| NVIDIA RT Cores | 110 |
| AI Performance | 2,142 AI TOPS |
| GPU Memory | 48 GB GDDR7 with ECC |
| Memory Interface | 384-bit |
| Memory Bandwidth | 1,344 GB/s |
| Max Power Consumption | 300W |
| Multi-Instance GPU | Up to 2x 24GB Up to 1x 48GB |
| Graphics Bus | PCI Express 5.0 x16 |
| Display Connectors | DP 2.1 (4) |
| Form Factor | 4.4" H x 10.5" L, FHFL Dual Slot |
| Product Weight | 1.219 kg |
| Thermal Solution | Blower Active Fan |
| Power Connector | 1x PCIe CEM5 16-pin |
| NVENC | NVDEC | JPEG | 3x | 3x | 1x |





















