The Coral M.2 Accelerator with Dual Edge TPU integrates two Edge TPUs into existing computer systems using an M.2 E-key interface.
This Edge TPU module is particularly well-suited for mobile and embedded systems that can benefit from accelerated machine learning.
Each Edge TPU coprocessor is capable of executing 4 trillion operations per second (4 TOPS) with a power consumption of 2 watts. For example, modern Mobile Vision models like MobileNet v2 can be efficiently executed at nearly 400 FPS.
With two Edge TPUs (and thus 8 TOPS), you can double the system's performance - for instance, by running two models in parallel or distributing the processing steps of a model across both Edge TPUs.
The M.2 Accelerator with Dual Edge TPU is supported on Debian Linux-based systems as well as Windows 10.
We also stock the Coral USB Accelerator - to retrofit systems that do not offer an M.2 interface with a USB 3.0 interface.
Technical Specifications Coral M.2 Accelerator with Dual Edge TPU
- 2 x Google Edge TPU ML accelerator coprocessor
- 8 TOPS (int8); 2 TOPS per Watt
Interface & Software Support
- Interface: M.2 E-key (with two PCIe Gen2 x1 lanes)
- Supports Linux and Windows 10 on the host system
- Dimensions: 22 mm x 30 mm x 2.8 mm (M.2-2230-D3-E)
- Weight: 2.5 g
- Operating Temperature: -40 to +85°C
Host System Requirements
- 64-bit Linux Debian 10.0 / Ubuntu 16.04 (or later)
- CPU Architecture: x86-64 or ARMv8 (64-bit)
- Windows 10 (64-bit)
- x86-64 CPU Architecture
Although the M.2 specification requires E-key sockets to provide two PCIe x1 lanes, most manufacturers only provide one PCIe. To use both Edge TPUs, please ensure that your socket indeed provides two PCIe x1 lanes.
Also, please refer to the datasheet for peak current requirements (up to 3 A per Edge TPU) and thermal management information. Each Edge TPU includes a built-in temperature sensor and allows configuring parameters for when it should be turned off.
Warning: System overheating can lead to fire or hardware damage!
Potential for Industrial Applications & Consulting
The Coral Edge TPU is a revolutionary product for machine learning applications! This enables embedded solutions that can, for example, detect defects in workpieces, recognize traffic situations, and much more.
We are happy to advise you on Google's Coral product range.
If interested, please contact us
Hardware & Software Development
We offer hardware and software development from a single source.
For example, Google provides a special module (SoM) in the Coral.AI project that can be easily designed into your hardware solution by us. Or let us design a complete board for your application.
We also offer software development for artificial intelligence. It is a young field with massive value potential for you and your company - please contact us if you are interested in this pioneering field.