DX-M1 AI Accelerator M.2 Module with 4GB LPDDR5 (25 TOPS)
- Delivery nationwide via The Courier Guy — R150 flat, free over R2,000
- Free collection from our Milnerton (Cape Town) warehouse
- Secure checkout — PayFast card payment or EFT
- Questions? WhatsApp us on +27 65 900 8570
Description
The DX-M1 AI Accelerator M.2 Module delivers 25 TOPS of INT8 inference performance within an efficient 2W to 5W power envelope. Built in a standard M.2 2280 form factor with 4GB LPDDR5 memory, it enables low-latency edge computing across ARM and x86 platforms. It is ideal for makers, educators, and engineers developing computer vision, visual SLAM, and robotics projects on hardware like Raspberry Pi 5 and LattePanda.
Server-class edge AI performance
The accelerator delivers 25 TOPS of INT8 processing power, enabling complex neural networks and multi-stream video analysis to run directly on-device without cloud connectivity. An energy-efficient architecture keeps power consumption between 2W and 5W, reducing heat build-up and preserving battery life in mobile robotic platforms.
High-bandwidth memory and flash storage
Equipped with 4GB of onboard LPDDR5 memory, the module efficiently manages large AI models and demanding batch workloads. An integrated 1Tbit QSPI NAND Flash memory provides stable firmware storage for reliable system operation.
Universal M.2 PCIe connectivity
Utilising the standard M.2 M-Key 2280 form factor, the module connects via PCIe Gen3 x4 and supports backward compatibility with x1 mode. This enables seamless plug-and-play expansion for x86 PCs, LattePanda single-board computers, and Raspberry Pi 5 systems using PCIe HATs.
Comprehensive DXNN development toolchain
The DXNN SDK streamlines the transition from model training to runtime execution on Ubuntu 20.04/22.04 LTS and Windows 10/11. It provides native support for PyTorch, TensorFlow, TensorFlow Lite, Keras, and XGBoost, allowing easy deployment through ONNX formatting.
Industrial reliability and temperature range
Engineered to withstand rigorous operating conditions, the module functions from -25°C to 85°C with thermal throttling and 25°C to 65°C non-throttling. Its robust physical profile and 3.3V operating voltage make it well suited for outdoor autonomous systems, industrial automation, and vision-guided robotics.
Video
Downloads & documentation
Specifications
| Processor Performance | 25 TOPS (INT8) |
| Interface | M.2 M-Key, PCI Express Gen3 x4 (compatible with x1 mode) |
| Memory | 4GB LPDDR5, 1Tbit QSPI NAND Flash |
| Power Consumption | 2W ~ 5W |
| Power Range | 3.3V±5% |
| Framework Support | PyTorch, ONNX, TensorFlow, TensorFlow Lite, Keras, XGBoost |
| Operating Systems | Windows 10/11, Ubuntu 20.04/22.04 LTS |
| Operating Temperature | -25°C ~ 85°C (Throttling); 25°C ~ 65°C (Non_ Throttling) |
| Product Dimensions | 22 mm x 80 mm x 4.1 mm/0.87 inch x 3.15 inch x 0.16 inch |
In the box
| DX-M1 AI Accelerator M.2 Module with 4GB LPDDR5 | 1 |
Reviews
No reviews yet — be the first to review this product.
From the same brand
Gravity: 4Pin I2C/UART Sensor Cable (10PCS)
Gravity: Analog CH4 Gas Sensor (MQ4) For Arduino
Gravity: Analog Gas Sensor (MQ2)
3xAAA Battery Holder
You might like
KY-008 650nm Laser Transmitter Module, 5V
Nano I/O Expansion Board for Arduino Nano V2.X and V3.X
1.3-Inch 128x64 Blue SPI OLED Display Module (SH1106)
