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 makes this Edge TPU module particularly well suited for mobile and embedded systems that can benefit from accelerated machine learning.

Each Edge TPU coprocessor is capable of performing 4 trillion operations per second (4 TOPS) with 2 watts of power consumption. For example, advanced mobile vision models such as MobileNet v2 can run efficiently at close to 400 FPS.

With two Edge TPUs (and thus 8 TOPS), you can double the performance of the system - for example, by running two models in parallel, or by distributing the processing steps of one model across both Edge TPUs ("pipelining").

The M.2 Accelerator with Dual Edge TPU is supported on Debian Linux-based systems as well as Windows 10 systems.

Please contact us for educational discounts & advice.

Tech Specs

Coral M.2 accelerator with Dual Edge TPU: The Coral M.2 Accelerator with Dual Edge TPU is an M.2 module that provides two Edge TPU coprocessors for existing systems with an available M.2 E-key slot.

TPU:

  • 2 x Google Edge TPU ML accelerator coprocessor
  • 8 TOPS (int8); 2 TOPS per Watt

Interface & software support:

  • Interface: M.2 E-key (via two PCIe Gen2 x1 lanes)
  • Supports Linux, and Windows 10 on the host system

Other information:

  • 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:

Linux:

  • 64-bit Linux Debian 10.0 / Ubuntu 16.04 (or newer)
  • CPU: x86-64, or ARMv8 (64-bit)

or Windows:

  • Windows 10 (64-bit),
  • x86-64 CPU

Important hints

Although the M.2 specification requires that E-key sockets provide two PCIe x1 lanes, most manufacturers only provide one PCIe interface! To be able to use both edge TPUs, please make sure that your socket actually provides two PCIe interfaces with x1 PCIe lane each.

Please refer to the datasheet for peak current requirements (up to 3 A per Edge TPU), as well as thermal management. Each Edge TPU contains a built-in temperature sensor and allows you to configure parameters when it should shut down.

Warning: Overheating of the system can lead to fire or destruction of hardware!

Potential for industrial applications

The Coral Edge TPU is a revolutionary product for machine learning applications! It enables embedded solutions that can, for example, detect problems with workpieces, recognize traffic situations, and much more.

Datasheet

Google part number:

  • G650-06076-01
Reseller information

Our Coral M.2 Accelerator with Dual Edge TPU Resellers (TBD)

Global Resellers

buyzero.de
thepihut.com
berrybase.de
Farnell (UK Site)

Germany, Austria and Switzerland

buyzero.de
welectron.com
berrybase.de
pollin.de
funk24.net
conrad.de
elektor.de
reichelt.de
shop.heise.de
Farnell (Deutschland)
Semaf.at (Österreich)

Denmark, Norway, Sweden and Finland (+ Baltic countries)

raspberrypi.dk

Netherlands and Belgium

raspberrystore.nl
elektor.nl

Poland

kamami.pl

Israel

Piitel.co.il (Israel)
Piitel.co.il (Israel)

Italy

Kubii.it

Spain and Portugal

tiendatec.es

United Kingdom and Ireland

thepihut.com

Additional Legal Information

The products offered by Google are unrelated to the products offered under the CORAL trademarks owned by Orient Development Enterprises Ltd. in Taiwan.