Professional

High-resolution eight DisplayPort or DVI monitor support Support resolutions up to 2560 x 1600 (DisplayPort) and 1920 x 1200 (DVI) per output Native PCI Express (PCIe) x16 performance Maximum bandwidth to ensure best performance 2 GB of memory For enhanced performance in demanding multi-monitor control room monitoring applications ATX form factor Stretched and independent desktop modes Expand your desktop across multiple monitors or set each monitor independent of each other Edge Overlap Easily and intuitively adjust the number of overlapping pixels to create a seamless, unified image on up to eight edge blending projectors. M-Series multi-card support1 Add an additional M-Series card into the same system to drive up to 16 displays from two cards Joined graphics card mode2 Insert two M-Series cards into the same system to drive even more monitors across a stretched desktop Matrox Clone View a copy of one display on the other displays Wide enterprise flexibility Unified driver package compatible with other Matrox products Matrox PowerDesk software Easily configure and manage program windows on multiple monitors Multiple operating system support

CUDA Parallel-Processing Cores 2560 GPU Memory 16GB GDDR5X FP32 Performance 8.9 TFLOPS Max Power Consumption 180W Graphics Bus PCI Express 3.0 x 16 Display Connectors DP 1.4 (4), DVI-D (1), Optional Stereo (1)

Four NVIDIA Maxwell GPUs Up to 32 users per board 2560 NVIDIA CUDA cores (640 per GPU) 32 GB of GDDR5 memory (8 per GPU) 28 H.264 1080p30 streams ECC protection for increased reliability Server optimization to deliver the best throughput in the data center

• CUDA Parallel-Processing Cores 4,608
• NVIDIA Tensor Cores 576
• GPU Memory 24 GB GDDR6
• RT Cores Yes
• Graphics Bus PCI Express 3.0 x 16
• NVLink Yes
• Display Connectors DP 1.4 (4), VirtualLink (1)
• Form Factor 4.4" (H) x 10.5" (L) Dual Slot

4992 NVIDIA CUDA cores with a dual-GPU design Up to 2.91 teraflops double-precision performance with NVIDIA GPU Boost Up to 8.73 teraflops single-precision performance with NVIDIA GPU Boost 24 GB of GDDR5 memory 480 GB/s aggregate memory bandwidth ECC protection for increased reliability Server-optimised to deliver the best throughput in the data center

• CUDA Parallel-Processing Cores 4,608
• NVIDIA Tensor Cores 576
• GPU Memory 48 GB GDDR6
• RT Cores Yes
• Graphics Bus PCI Express 3.0 x 16
• NVLink Yes

• CUDA Parallel-Processing Cores 3840
• GPU Memory 24GB GDDR5X
• FP32 Performance 12 TFLOPS
• Max Power Consumption 250 W
• Graphics Bus PCI Express 3.0 x 16
• Display Connectors DP 1.4 (4), DVI-D (1), Optional Stereo (1)

CUDA Parallel-Processing Cores 5120 NVIDIA Tensor Cores 640 FP64 Performance 7.4 TFLO FP32 Performance 14.8 TFLOPS FP16 Performance 29.6 TFLOPS Tensor Performance 118.5 TFLOPS Max Power Consumption 250 W Graphics Bus PCI Express 3.0 x 16 Display Connectors DP 1.4 (4) Form Factor 4.4 H x 10.5 L Dual Slot

140X HIGHER THROUGHPUT TO KEEP UP WITH EXPLODING DATA SIMPLIFIED OPERATIONS WITH A SINGLE TRAINING AND INFERENCE PLATFORM REAL-TIME INFERENCE FASTER DEPLOYMENT WITH NVIDIA DEEP LEARNING SDK

GPU Architecture NVIDIA Volta NVIDIA Tensor Cores 640 Double-Precision Performance 7 TFLOPS Single-Precision Performance 14 TFLOPS Tensor Performance 112 TFLOPS GPU Memory 16GB HBM2