The Coral Dev Board is a single-board computer with a removable system-on-module (SOM) that contains eMMC, SOC, wireless radios, and Google’s Edge TPU. It’s perfect for IoT devices and other embedded systems that demand fast on-device ML inferencing.

You can use the Dev Board as a single-board computer for accelerated ML processing in a small form factor, or as an evaluation kit for the SOM that’s on-board. The 40 mm × 48 mm SOM can be combined with your custom PCB hardware using board-to-board connectors for integration into products.

The SOM is based on NXP's iMX8M system-on-chip (SOC), but its unique power comes from the Edge TPU coprocessor. The Edge TPU is a small ASIC designed by Google that provides high performance ML inferencing with a low power cost. For example, it can execute state-of-the-art mobile vision models such as MobileNet v2 at 100+ fps, in a power efficient manner.

Edge TPU Key Benefits

   - High speed TensorFlow Lite inferencing

   - Low power

    - Small footprint

The baseboard includes all the peripheral connections you need to prototype a project, including USB 2.0/3.0 ports, DSI display interface, CSI2 camera interface, Ethernet port, speaker terminals, and a 40-pin GPIO header.

Coral by Google is a platform for building devices with local AI

Features

SOM

    - NXP i.MX 8M SOC (Quad-core Cortex-A53, plus

    - Cortex-M4F)

    - Google Edge TPU ML accelerator coprocessor

    - Cryptographic coprocessor

    - Wi-Fi 2x2 MIMO (802.11b/g/n/ac 2.4/5GHz)

    - Bluetooth 4.1

    - 8GB eMMC

    - 1GB LPDDR4

USB Connections

    - USB Type-C power port (5V DC)

    - USB 3.0 Type-C OTG port

    - USB 3.0 Type-A host port

    - USB 2.0 Micro-B serial console port

Audio connections

    - 3.5mm audio jack (CTIA compliant)

    - Digital PDM microphone (x2)

    - 2.54mm 4-pin terminal for stereo speakers

Video connections

    - HDMI 2.0a (full size)

    - 39-pin FFC connector for MIPI-DSI display (4-lane)

    - 4-pin FFC connector for MIPI-CSI2 camera (4-lane)

MicroSD card slot

Gigabit Ethernet port

40-pin GPIO expansion header

Supports Debian Linux

Models are built using TensorFlow

Our compiler fully supports MobileNet and Inception model architectures though custom architectures are possible.