NVIDIA Jetson Nano Developer Kit (945-13450-0000-100)

Very good shipping, everything arrived well, and the quality of the board is very good, as well as the efficiency to process ML algorithms

The NVIDIA Jetson Nano is a compact, affordable AI development board designed for makers, developers, students and robotics enthusiasts. Despite its small size, it packs impressive performance for real-time computer vision and deep learning tasks.⸻Specifications• CPU: Quad-core ARM Cortex-A57 @ 1.43 GHz• GPU: 128-core NVIDIA Maxwell• RAM: 4 GB LPDDR4• Storage: microSD card slot• Connectivity: Gigabit Ethernet (Wi-Fi via USB adapter)• Ports: 4x USB 3.0, HDMI, DisplayPort, MIPI CSI camera, GPIO header• Power: 5V/4A via barrel jack or micro-USB⸻Pros• Affordable – Great entry point into edge AI at under $400• Good Performance – Can run multiple neural networks for image classification, object detection, etc.• Strong Community Support – Tons of tutorials, forums, and projects available• Flexible I/O – Easy to connect cameras, sensors, and other hardware⸻Cons• Only 4 GB RAM – Can be limiting for heavier models or multitasking• No Built-in Wi-Fi/Bluetooth – External adapter needed• Older OS by Default – Ships with Ubuntu 18.04, which might require tweaks for newer packages• Power Supply Not Included – Must be purchased separately⸻Best Use Cases• AI and ML learning projects• Autonomous robotics and object detection• Real-time edge computing and smart vision systems⸻ConclusionThe Jetson Nano is one of the best budget-friendly AI platforms available. It’s perfect for prototyping, learning AI, and building robotics or vision-based applications. If you need more power, you might consider the Jetson Orin Nano—but for most starter projects, the Nano is a fantastic choice.

Jetson Nano is great for not only robotics/edge AI, you can use ML for science usage such as medical imaging or environmental data to speed up your workflow. From personal experience even an underclocked dual-core power save mode on the Jetson Nano will still be faster on CUDA AI/ML tasks than a Raspberry Pi 4, however your workflow may vary. If you use AI/ML that isn't optimized for CUDA, in some cases a Pi 4 raw CPU compute can edge out the Nano. I would say if you pair a Pi 4 with any AI/ML accelerator it'll cost more than a Jetson Nano and your mileage is still going to vary.Depending upon how you use a Jetson Nano, for robotics/automation you can actually run four cameras via USB and use the camera interface. Performance wise if you do opt to run a Jetson Nano using USB power, your mileage is going to vary as not all USB power adapters provide a stable voltage which means checking the specs--I reused a Canakit USB power adapter from a retired Pi 3 and never had any voltage warnings but if you plan to run a Jetson Nano hard like a Pi 4 you'll want to use the barrel power adapter for extra power stability when using multiple USB devices+GPIO. Thermal wise I've compared a fanless vs fan equipped Jetson Nano, even under sustained load the heatsink size prevents it from thermal throttling too much. This B01 version has two camera connectors which is geared for stereo imaging however you can run two cameras at a small performance loss and also fixed the networking issue which occurred on the original Jetson Nano A01/A02.From a performance per watt/dollar ratio, if you're going to dive deeper into AI/ML a Jetson NX is more ideal. With a Jetson Nano if you're pushing four cameras and LIDAR it'll require a bit of tweaking to get optimal performance and still remain at about 3.5GB of memory usage.

Amazing product

I expected this to be, basically, a Raspberry Pi with a GPU. A computer I could flash with any arm64/aarch64 OS and it Just Work. Boy, was I wrong.Despite being called a "Dev Kit", this is just the device itself. Nvidia just calls it a " non-production-grade" product, so Dev Kit is just a euphemism for it being a low-quality product, on purpose.The docs say you need a 5V 2A power supply, but that is not true. It will not even boot unless it has at least 4A, so make sure you buy the right one.The Jetson Linux SD card image in the Getting Started docs is half a decade old. Even the most current version of Jetson Linux available for this product is based on Ubuntu 18.04, which reaches EOL in May 2023. Nvidia is *not* releasing updates for Jetson Linux that are compatible with this product, so it will be obsolete and wholly unsupported by both Nvidia and Canonical after May 2023. So you're on your own.Even after getting Jetson Linux up and running, it's stuffed with bloatware like OS-integrated Facebook search (circa 2018) and a variety of games and a vastly outdated version of the whole Libreoffice suite. The default desktop theme and wallpaper is nausea-inducingly garish, reminiscent of the fad around lightning and dragon themed bowling shirts, circa 1998.It is possible to upgrade the OS from Ubuntu 18.04 all the way up to 21.04 if you're motivated enough. Each upgrade causes more volatility and instability, though, so YMMV and if you're planning to do this, be very cautious and make backups along the way. I spent 7 hours attending this process (yes, you have to attend it because along the way, there are random confirmations you actually have to read, because sometimes the default is to abort the process, setting you back an hour or more. Ask me how I know this). After carefully upgrading, one major version at a time (bionic to focal, focal to jammy), the upgrade from jammy to koala bricked the OS and none of my recovery attempts were fruitful.The problem with just using Jetson Linux right out of the box, without upgrading, is that you have to work within the time capsule of what was available at that time. Modern tools, AI models, etc, developed since will either not have the necessary dependencies, or taken as a whole, will not work with such an antiquated system. Combine that with this product being effectively abandoned with no future in sight for official support or continued development, and even if you're a seasoned, highly competent software engineer, you're in for a world of hurt.If you do buy this product, I hope you have a lot of patience and tolerance for disorder and chaos. You're going to need all of it.

para empezar muy bueno, sobre todo en procesamiento de imagenes