Datasheet: SBCs On the Edge of AI - To reduce latency, single-board computers take inferencing closer to the source

A diversity of missions, size, computing resources and power consumption give designers a range of options for choosing the right board to fit the application.

- Tom Murphy

Originally designed to be low-cost, compact, fanless marvels of power efficiency and low heat, single-board computers are now carrying the AI vanguard toward edge computing applications.

It's all about making inferences where they are needed rather than sending them to the core or data center, where latency might negate any advantages.

Finding a common thread among a batch of SBC products can be difficult because there is a diversity of missions, sizes, computing resources, and efficiency options depending on what the devices' use, where they will be deployed, or the applications required. But AI, inferencing, NPUs, and TOPS are making waves in the segment.