- Out-of-Stock
- Motorized
- Tools
-
Dev. Tools
-
Electronics
- Power and Energy
- Wires and Connectors
- Grove
- USB Peripherals
- Motors and Servos
- LEDs & Lights
- Assembly kits
- Domotic
- Break-out Boards
- Network
- Electronic basics
- PCBs
- Sound
- Integrated Circuits (ICs, Chips)
- Wireless Connectivity
- Sensors
- Human Interface
- CMS / SMD
- Others
- Prototyping
Google Coral M.2 Accelerator (A+E key)
€36.00
Tax included
M.2 Accelerator module (A+E key) integrating Google's Edge TPU coprocessor for edge machine learning inference.
Delivering 4 TOPS at 2 TOPS/Watt, it is compatible with Debian Linux and Windows 10, supporting TensorFlow Lite models compiled for hardware acceleration.
The Google Coral M.2 Accelerator is a compact module that integrates Google's Edge TPU coprocessor into any system with an available M.2 slot (A+E key). It enables high-speed machine learning inference at the edge, reducing reliance on cloud processing for AI workloads.
This accelerator is compatible with Debian-based Linux systems and Windows 10. It supports TensorFlow Lite models compiled specifically for the Edge TPU, allowing state-of-the-art computer vision models (such as MobileNet v2) to run efficiently with low power consumption.
Specifications
- ML Accelerator: Google Edge TPU coprocessor
- Performance: 4 TOPS (int8), delivering 2 TOPS per watt
- Interface: M.2 A+E key (PCIe)
- Form Factor: 22 mm x 30 mm (M.2-2230-A-E-S3)
- OS Compatibility: Debian-based Linux, Windows 10
- Framework Support: TensorFlow Lite (compiled via Edge TPU Compiler)
For further technical details, visit the official Coral website.
ACCMODCORAL
You might also like
M.2 Accelerator module (A+E key) integrating Google's Edge TPU coprocessor for edge machine learning inference.
Delivering 4 TOPS at 2 TOPS/Watt, it is compatible with Debian Linux and Windows 10, supporting TensorFlow Lite models compiled for hardware acceleration.