Telemonitoring module RTX4080m MXM

Product overview
MM4080B6-12G is an MXM 3.1 Type B graphics module based on the NVIDIA Ada Lovelace architecture. It integrates an NVIDIA GeForce RTX4080 Max-Q GPU with 12GB GDDR6 (192-bit), 7427 CUDA cores and supports PCIe 4.0 x16 and multiple 8K-capable outputs for high-end embedded and professional systems.

Key features

• NVIDIA Ada Lovelace architecture — GeForce RTX4080 Max-Q
• 12GB GDDR6, 192-bit memory, 432 GB/s bandwidth
• 7427 CUDA cores; 232 Tensor cores; 80 RT cores
• MXM 3.1 Type B (105 × 82 mm), PCIe 4.0 x16 interface
• Display outputs: 2× DisplayPort 1.4++, HDMI 2.1, eDP, HDMI (multi-display); DP up to 8K@120Hz
• TDP 125W (optional 80W); DC input 12–19V; supports Windows 10/11 64-bit and Linux

Performance details
Peak single-precision (FP32) performance ~30 TFLOPS. The combination of a high-bandwidth GDDR6 memory subsystem and PCIe 4.0 interface enables low-latency, high-throughput workloads such as high-resolution rendering, real-time ray tracing and AI inference on edge and embedded platforms.

Technical specifications

• Model: MM4080B6-12G (RTX4080m)
• GPU: NVIDIA GeForce RTX4080 Max-Q (Ada Lovelace)
• Memory: 12GB GDDR6, 192-bit, 432 GB/s
• Compute: 7427 CUDA cores; 232 Tensor cores; 80 RT cores; 30.0 TFLOPS (FP32)
• Interface: MXM 3.1 Type B (105 × 82 mm), PCIe 4.0 x16
• Outputs: 2× DP 1.4++, HDMI 2.1, eDP, HDMI
• Max resolution: DP up to 7680×4320@120Hz; HDMI up to 8K@60Hz
• Power: TDP 125W (optional 80W); Input DC 12–19V; 3.3V & 5V ±5%
• OS support: Windows 10/11 64-bit, Linux
• Operating temp: 0°C to 45°C; Storage: -20°C to 75°C; Humidity ranges as specified

Typical applications

• High-performance embedded and industrial computing
• Professional graphics workstations and visualization
• Edge AI inference and machine learning acceleration
• Simulation, rendering and video wall systems