PCI Express graphics card MMR5000B6-16G

Product introduction
MMR5000B6-16G V2 is an MXM 3.1 Type B GPU module based on the NVIDIA Ada Lovelace architecture, equipped with the NVIDIA Quadro RTX5000 Mobility GPU. It integrates 9,728 CUDA cores, 304 fourth‑generation Tensor Cores and 76 third‑generation RT Cores, delivering up to 37.0 TFLOPS (FP32). The module provides 16GB GDDR6 (256‑bit) memory with 576GB/s bandwidth.

The module measures 105 × 82 mm and connects via PCIe 4.0 x16. It offers 4 × DisplayPort 1.4++ outputs and supports single‑display resolutions up to 7680×4320@120Hz. Supported operating systems include Windows 10/11 64‑bit and Linux. Nominal TDP is 115W, with optional 60W/80W configurations for low‑power deployments. Operating temperature range is 0°C to 45°C, suitable for embedded and edge computing applications.

Key features

• NVIDIA Ada Lovelace architecture with NVIDIA Quadro RTX5000 Mobility GPU
• 16GB GDDR6, 256‑bit memory, 576GB/s bandwidth
• 9,728 CUDA cores; 304 Tensor Cores; 76 RT Cores
• MXM 3.1 Type B form factor (105×82 mm); PCIe 4.0 x16 interface
• 4× DisplayPort 1.4++ outputs; up to 8K@120Hz
• Configurable TDP: 115W (options 60W/80W)
• Compatible with Windows 10/11 64‑bit and Linux

Technical specifications

• Model: MMR5000B6-16G V2 (RTX5000m)
• GPU architecture: NVIDIA Ada Lovelace
• GPU: NVIDIA Quadro RTX5000 Mobility
• Memory: 16GB GDDR6, 256‑bit
• Memory bandwidth: 576GB/s
• Core clocks (Base/Boost): 1425 / 2115 MHz
• CUDA cores: 9,728
• Tensor Cores: 304 (4th gen)
• RT Cores: 76 (3rd gen)
• TMUs: 304
• ROPs: 112
• SM count: 76
• Interface: PCI Express 4.0 x16
• MXM standard: MXM 3.1 Type B
• Dimensions: 105 × 82 mm
• TDP: 115W (configurable 60W / 80W)
• Display outputs: 4 × DisplayPort 1.4++
• Max resolution: 7680×4320@120Hz
• Operating systems: Windows 10/11 64‑bit, Linux
• Operating temperature: 0°C ~ 45°C
• Storage temperature: -20°C ~ 75°C
• Operating humidity: 0% ~ 95% (non‑condensing)
• Storage humidity: 10% ~ 90%
• Supply voltage: DC 12~19V, 3.3V & 5V ±5%
• FP32 performance: 37.0 TFLOPS
• FP64 performance: 691.6 GFLOPS
• DirectX: 12 Ultimate (12_2)
• OpenGL: 4.6
• OpenCL: 3.0
• Vulkan: 1.3
• Shader Model: 6.7
• CUDA version: 8.9