The LattePanda Mu A380 MXM GPU Carrier Board Kit is positioned as an ultra-compact, modular x86 edge computing platform that bridges the gap between low-power single-board computers (SBCs) and bulky industrial PCIe expansion systems. It offers a unique combination of scalable x86 processing, modular MXM discrete graphics, and 6-channel 4K video output within a standardized embedded footprint.

Note: Compute Module Required, the LattePanda Mu Compute Module is NOT included. Please purchase it separately to complete the system.

What Challenges Does the LattePanda Mu A380 MXM GPU Kit Solve?

Standard x86 Single Board Computers (SBCs) frequently bottleneck when deployed in high-density visual applications or localized artificial intelligence pipelines. Integrated graphics lack the hardware decoding capabilities and parallel processing cores required for complex matrix layouts and real-time neural network inference.

The LattePanda Mu A380 MXM GPU Carrier Board Kit bridges this gap by marrying the modular efficiency of the LattePanda Mu compute module with a discrete MXM graphics architecture. Housed in a highly integrated footprint, this solution provides desktop-class graphical capabilities and independent multi-screen display management without the spatial or thermal overhead of standard PCIe expansion chassis.

How Does the Modular Architecture Enhance Edge Deployments?

This carrier board kit utilizes a decoupled processing architecture that separates the CPU compute layer from the GPU acceleration subsystem. This design allows developers to scale system components independently based on strict deployment budgets or thermal envelopes.

Decoupled Scaling: The core carrier slot accepts the full range of LattePanda Mu compute modules, allowing seamless interchangeability between power-efficient processors (such as the Intel N100) and higher-compute alternatives (such as the Core i3-N305).

Interchangeable MXM Graphics: Compliance with the MXM 3.0/3.1 (Type A/B) standards guarantees flexible GPU hardware tailoring. The system scales effortlessly from low-power nodes to high-compute configurations depending on the workload matrix.

Two Powerful GPU Kit Options

Two distinct versions of the MXM GPU Carrier Board Kit are available to meet diverse computing requirements. This page details the Intel Arc A380 Kit, optimized for efficient, ultra-low-power edge deployments.

Targeting Ultra-Low-Power Edge AI (Intel Arc A380 / 6GB VRAM): Equipped with the Intel Arc A380, this system transforms into a highly cost-effective, ultra-low-power edge AI node. Perfectly tailored for streamlined tasks, the configuration effortlessly runs INT4-quantized Large Language Models (LLMs) at ~3B parameters locally. It stands as the ideal choice for entry-level local AI projects, high-density digital signage, and multi-display setups that require excellent hardware decoding performance within a strict power and thermal budget.

For projects demanding heavier compute power and massive VRAM for advanced AI workloads, please refer to the RTX 3060 MXM GPU Carrier Board Kit.

Modular CPU Configuration

Fully compatible with all the LattePanda Mu compute modules. Drop in either the Intel N100 or Core i3-N305 processor based on your deployment requirements. This decoupled architecture allows you to scale CPU compute independently from the MXM GPU selection.

Important Note:

• Power Safety: NEVER connect the DC Jack and the ATX 8-Pin power supply simultaneously. This will cause permanent damage to the board.
• Fan Voltage: Check your MXM fan voltage (5V or 12V) and set the PWR_FAN jumper accordingly before powering on to avoid overheating or fan failure.

Specifications:

Core Slot

• Compute Module: LattePanda Mu Slot
• GPU Module: MXM 3.0/3.1 (Type A/B Compatible)

Video Output

• HDMI: 3 x HDMI, Each supports 4K@60Hz (Mu iGPU 2-ch + MXM dGPU 1-ch)
• DisplayPort(DP): 3 x DP, Each supports 4K@60Hz (All from MXM dGPU)

Expansion Slots

• M.2 M Key: 2230, PCIe 3.0 x1
• M.2 E Key: 2230, PCIe 3.0 x1 + USB2.0

Internal I/O

• F_Front: PWRBTN, RESET, PLED (Standard 2.54mm Pitch)
• CPU Fan: PH2.0-4P
• MXM Fan: PH2.0-4P & PH2.5-4P
• PWR_FAN Jumper: 2.54mm Pin Header (5V/12V selectable)
• RTC: CR1220 battery holder

External Ports

• USB: 2 x USB 3.2 Gen2 + 2 x USB 2.0
• Ethernet: 1 x RJ45, 10/100/1000Mbps Auto-Negotiation

Mechanical

• Carrier Board Size: 146 x 102mm
• Overall Size: 159 x 107.5 x 88.5mm

Applications:

• High-Density Digital Signage & Video Walls
• Localized Edge AI Inference
• Interactive Media Installations
• Multi-Stream Industrial Machine Vision

Documentation:

• Product Wiki

Package Includes:

• 1 x LattePanda Mu GPU Carrier Board
• 1 x Intel Arc A380 MXM Graphics Card (Pre-assembled with cooling fan)
• 1 x Metal Enclosure
• 1 x Power Adapter
• 3 x AC Power Cords (US/EU/UK standards included)
• 1 x 12mm Metal Push-button Switch (with integrated LED indicator)
• 1 x Screw Pack
• 1 x CR1220 RTC Battery
• 1 x Thermal Pad (10x10x5mm)