Language: EN


Coral Dev Board Micro, Google's AI board

The Coral Dev Board Micro is a development board created by Google, as part of its devices for artificial intelligence and edge computing.

This board is a more compact and low-power version than its predecessors, the original Coral Dev Board and the Coral Dev Board Mini, and is designed to provide developers with an accessible tool to create AI solutions at the edge.

The Coral Dev Board Micro is designed to be used in edge AI solutions, such as in IoT devices, robots, and security systems. With its Edge TPU coprocessor, the board can run AI models in real time with extremely low power consumption.

The Coral Dev Board Micro uses an NXP i.MX RT1176 Cortex-M7/M4 processor and a Google Coral Edge TPU coprocessor, accompanied by 128MB of flash memory and 64MB of RAM.

In addition, the board includes a 324x324px camera and a microphone, making it very suitable for vision and audio applications. For example, we could use it for a voice recognition system project or to analyze images in real time.


Its specifications include:

  • NXP i.MX RT1176 processor with 2 Arm Cortex-M7 and Cortex-M4 cores
  • Coral Edge TPU coprocessor (2 TOPS/Watt)
  • 128MB flash memory
  • 64MB RAM
  • 2MB SRAM
  • 324x324 resolution Himax camera
  • Mono PDM microphone
  • 1x USB 2.0 Type-C port
  • Optional WiFi 5 and Bluetooth 5.0 connectivity
  • 2x 12-pin GPIO connectors
  • 2x 100-pin connector Coral board-to-board
  • 4 LEDs
  • 2 buttons
  • Dimensions 65x30mm

The Coral Dev Board Micro is really small, especially when compared to the rest of the coral family. At 65x30mm, it is similar in size to a Raspberry Pi Zero.

In terms of connections, it has 2 rows of 12 pins for GPIO, 4 LEDs, and two buttons. In addition to a 100-pin connector (!) intended for connecting multiple boards together. For power and programming, there is a USB-C port.

Regarding software and programming, the board comes standard with FreeRTOS, which is the recommended way to program the Coral Dev Board Micro. However, it is also possible to program it with the Arduino environment.

Finally, in terms of AI capabilities, the Coral Dev Board Micro is compatible with TensorFlow Lite for Microcontrollers (TFLM). In addition, Google provides extensive documentation and tutorial guides to help developers get started with using the board.

The Coral Dev Board Micro is not yet available, but it can be pre-ordered at a price of $79.99. More information and all the documentation on the product page at