NXP FRDM-IMX95 - Unboxing | DigiKey

We take a look at the FRDM i.MX 95 development board from NXP, a low-cost and compact board suitable for advanced HMIs, industrial networking, vision systems, and advanced edge AI applications. Inside the box, you’ll find the FRDM i.MX 95 development board with an M.2 Wi-Fi and Bluetooth module based on NXP’s IW612, a USB-A to USB-C cable, a USB-C to USB-C cable, and some standoffs. Also note there is a second M.2 slot that can be used with an AI accelerator card. The board is pre-programmed with a Linux BSP image with NXP GoPoint demos. You can follow NXP’s getting started guide to make sure you’re on the latest version. Some of the demos require an internet connection. We’ll use ethernet, but you can also setup Wi-Fi. Connect the board’s HDMI output to a monitor. Connect your computer to the board’s USB debug port. Before connecting to power, setup terminals for the four UART connections the board provides to debug the Arm Cortex-A55, M33, and M7. Then connect a mouse to the USB-A port and plug in a USB-PD source to the USB-C power input and slide the switch to power on the board. You’ll see output on three of the terminals and can login to the console, but we’ll continue on the GUI. Open GoPoint and you’ll see a number of demos. ML Benchmark allows us to compare the A55 CPU inferencing performance to the integrated eIQ Neutron NPU. You can see that NPU-inferencing is significantly faster. This makes ML inferencing more responsive and more efficient, both of which are critical for edge AI devices to be successful.