Pleasanton, Calif. – September 12, 2026 -- Alif Semiconductor launched a new family of low-cost evaluation boards priced at an MSRP of $49, aimed at lowering the barrier to on-device AI/ML inference for developers.
StartKit boards target two MCU families for edge AI development
The SK-E1C board supports Alif's Ensemble microcontrollers, while the SK-B1 targets the BLE wireless Balletto family. Both boards ship with a superset configuration combining a 160MHz Arm Cortex-M55 CPU core with Helium vector processing extension, an Ethos-U55 NPU, and 2MB of tightly coupled SRAM and MRAM.
NPU delivers inference speeds up to two orders of magnitude faster than traditional MCUs
Alif states the Ethos-U55 NPU included in both boards accelerates inference performance by up to 100x compared to running identical workloads on a conventional MCU core, positioning the StartKit line for battery-powered, always-on AI applications.
Boards support Arduino and MikroE Click expansion standards
Beyond Alif's own 44-pin dual-row expansion headers exposing all MCU functions, the StartKit platform accommodates Arduino R3 shields and MikroE Click add-ons for sensors and audio. Each unit includes a micro-B USB port for data, USB-C for power, a tri-color LED, a user button, and current-sense jumpers for power profiling. The SK-B1 additionally integrates a tuned chip antenna and RF filter network for BLE connectivity.
SEGGER J-Link debugger is built into the board for streamlined development
Alif integrated SEGGER's J-Link On-Board debugger directly into the StartKit platform to simplify development and debugging workflows for engineers evaluating the hardware.
Mark Rootz, VP of Marketing at Alif Semiconductor, said the company introduced the first NPU-enabled microcontrollers in 2021 and that the StartKit platform now extends that technology to a broader base of developers.