NVIDIA came up with a cheap alternative to that of the other SBCs we saw previously called Jetson Nano. This is a $99 board that is small and a powerful computer that lets us run multiple neural networks in parallel for applications such as image classification, object detection, segmentation, and speech processing. It comes with a 128-core Maxwell GPU and a Quad-Core ARM A57 @ 1.43 GHz CPU, with 4 GB 64-bit LPDDR4 25.6 GB/s memory. There is no eMMC but it comes with an SD card slot for storage or to run an OS. It has Gigabit Ethernet and HDMI 2.0 and eDP 1.4 display ports, and it supports I2C, I2S, SPI, and UART. It runs at as low as 5 watt power. Jetson Nano is shown here:

Now, let's look at a comparison between Jetson Nano and Jetson TX2.