Back to DFRobot

DFRobot DX-M1 AI Accelerator M.2 Module

DFRobot DX-M1 AI Accelerator M.2 Module delivers server-class edge computing power with 25 TOPS (INT8) of inference performance in a standard M.2 2280 form factor. The DX-M1 offers 4GB LPDDR5 memory and a PCIe Gen3 x4 interface, allowing low-latency processing and seamless compatibility with Raspberry Pi 5, LattePanda, and various x86/ARM platforms.

The DFRobot DX-M1 AI Accelerator M.2 Module is supported by the comprehensive DXNN® SDK for PyTorch, ONNX, and TensorFlow, making it an ideal solution for placing complex AI models in intelligent robotics, visual SLAM, and industrial automation systems.

Features

  • AI accelerator
  • M.2 M key (22mm x 80mm) form factor
  • PCle Gen.3 x4 interface
  • 4GB LPDDR5, QSPI 1Gbit NAND Flash
  • Host HW: x86, ARM based architecture

Applications

  • Robotics: visual SLAM and navigation
  • Autonomous systems: drone perception and self-driving platforms
  • Industrial: AI visual inspection and safety monitoring
  • Edge computing: real-time video analytics and object detection

Specifications

  • 25 TOPS (INT8) processor performance
  • M.2 M-Key, PCI Express Gen3 x4 interface
  • 4GB LPDDR5, 1Tbit QSPI NAND Flash
  • 2W to 5W power consumption
  • Framework support: PyTorch, ONNX, TensorFlow, TensorFlow Lite, Keras, XGBoost
  • 3.3V±5% power range
  • Windows 10/11, Ubuntu 20.04/22.04 LTS operating systems
  • -25°C ~ +85°C (throttling); 25°C ~ 65°C (non_ throttling) operating temperature
  • 22mm x 80mm x 4.1mm/0.87" x 3.15" x 0.16" product dimensions

Videos

Functional Block Diagram

Block Diagram - DFRobot DX-M1 AI Accelerator M.2 Module

Layout

DFRobot DX-M1 AI Accelerator M.2 Module
Published: 2026-05-06 | Updated: 2026-05-06