Google's new mini SBC, the Coral Dev Board Mini, was designed with machine learning in mind. This board was first announced back in 2019 with plans for an early 2020 release. As of October 2020, it's finally about to reach end users' hands. The Dev Board Mini is listed on Google's Coral store and on Seeed Studio, where, as HotHardware reports, it was previously available for pre-order with a ship date of October 31st (now it's showing as out-of-stock).
This new SBC has a 40-pin connector intended to interface with the previously released TPU accelerator module.
It runs a Linux derivative of Debian known as Mendel, opening support to many Linux tools and applications. It also supports AutoML Vision Edge.
The new board features a MediaTek 8167s SoC (Quad-core Arm Cortex-A35) processor with integrated graphics using IMG PowerVR GE8300. It has 2GB of LPDDR3 for memory.
It supports both Bluetooth and wireless connectivity. You can connect audio devices using a 3.5mm audio jack or output video using the micro HDMI port. A camera can be connected using a 24-pin FFC connector.
The Coral Dev Board Mini is positioned to compete against Nvidia's Jetson Nano which is available with 2GB for $59 and 4GB for $99. It's also, obviously, in some competition with the Raspberry Pi, which can do basic machine learning such as facial recognition (see how to train Raspberry Pi for facial recognition). The Pi can also get a performance boost from Google's own Coral USB accelerator dongle, which costs $59.
The Coral Dev Board Mini will be available soon. In the meantime, you can check it out on the official Coral website. It's currently listed for $99.
Stay on the Cutting Edge
Join the experts who read Tom's Hardware for the inside track on enthusiast PC tech news — and have for over 25 years. We'll send breaking news and in-depth reviews of CPUs, GPUs, AI, maker hardware and more straight to your inbox.
Ash Hill is a Freelance News and Features Writer at Tom's Hardware US. She manages the Pi projects of the month and much of our daily Raspberry Pi reporting.