GAP8 is a RISC-V based MCU the enables devices to interpret images, sounds and vibrations and last for years on batteries. GAP8 combines the attributes of an ultra-low power MCU with an 8 core compute cluster that delivers a 20 times energy efficiency improvement versus available solutions. GAP8 is not just a neutral network accelerator, it is a fully programmable device that can be used to optimize a range of signal processing and machine learning algorithms. It's target markets are people and object counting applications, image interpretation, wearables, security applications and fault detection and analysis.
GAP8 is based on work from the open source PULP project which itself leverages the open source RISC-V instruction set architecture.