Nvidia's $99 Jetson Nano Developer Kit brings GPU-supercharged AI smarts to maker projects

Look at the 128-core brain on Jetson!

Machine learning is coming to the masses, and those hordes of DIY drones and robots are about to get a whole lot smarter. On Monday at Nvidia’s GTC conference, the company plans to reveal the $99 Jetson Nano Developer Kit. The kit is an expansion of the company’s “Jetson” embedded graphics platform, and it aims to infuse your wildest maker projects with AI that the Raspberry Pi could only dream of. It’ll be available immediately online, through distributors, and at GTC itself.

The Jetson Nano Developer Kit is a standalone version of the new Jetson Nano AI computer also announced today. As you’d expect from an Nvidia product, the GPU’s the secret sauce—a 128-core beast that looms large on the maker board. Based on the “Maxwell” GPU architecture that powered the company’s GeForce GTX 900-series graphics cards, it’s capable of 472 GFLOPS of FP16 performance, HEVC video encoding at either 4K30 or 3x 1080p30 resolution, and HEVC decoding at 4K60 or 8x 1080p30—meaning it can monitor several high-definition video inputs at once and process the images it sees in real time. This is a powerful GPU to hand to makers. It’s tamed by a massive black heatsink sitting atop the module.

dsc00535 Brad Chacos/IDG

Click to enlarge

Nvidia fleshed out the board with a 1.43GHz quad-core ARM A57 processor, 4GB of 64-bit LPDDR4 memory, and gigabit ethernet. Storage is handled via a bring-your-own SD card, just like with the Raspberry Pi. Also like the Raspberry Pi, the Jetson Nano Developer Kit is loaded with hardware connections designed to expose the board’s smarts to the world. Nvidia equipped the board with three USB 2.0 ports, a USB 3.0 port, a MIPI-CSI camera connector, HDMI 2.0, DisplayPort 1.2, and I2S audio out, along with expansion options for a Wi-Fi card, a fan, and a UART header.  

You’ll also find a 40-pin GPIO header. Nvidia says the Jetson Nano Developer kit will support the Blinka library and thus many Adafruit peripherals. Instructions explaining how to configure the header to work with other add-on hardware can be found in the Jetson GPIO Python library.

jetson nano dev kit Nvidia

As with many other maker boards, the Jetson Nano Developer kit doesn’t ship with a power supply, so you’ll need to bring your own standard 5V/2A-3.5A micro-USB power charger. If you plan on running particularly intense benchmarks or workloads, Nvidia recommends using a 5V/4A DC power supply instead. The board can be configured to operate in either 5W or 10W mode depending on the intensity of your tasks.  

Of course, hardware is only half of the equation. That fancy GPU can’t do much without software to tell it what to do.

Jetson supports Linux4Tegra, an operating system derived from Ubuntu 18.04. It also supports popular deep learning software frameworks like TensorFlow, PyTorch, Caffe, Keras, and MxNet across a wide variety of models. The board’s hefty hardware help it run a wide range of AI frameworks and neural networks much faster than competing solutions, according to Nvidia, including a Raspberry Pi equipped with Intel’s Compute Stick.

jetson ai Nvidia

The Jetson Nano Developer Kit also supports Jetpack, Nvidia’s regularly updated software development kit that runs across the entire Jetson embedded products portfolio. Nvidia has been developing Jetpack for over three years at this point. The company also created a pair of tutorials dubbed “Hello AI World” and “Two Days to a Demo” that explain different inferencing and training approaches for makers who aren’t used to deep learning.

Finally, Nvidia developed an open-source “JetBot” AI project that shows off the Jetson Nano’s capabilities. Here’s how the reviewer’s guide describes the project:

“Through the ​J​etBot project makers and learners can get the hands-on experience needed to create innovative and fun AI projects with Nvidia Jetson Nano.  The project achieves this by taking the user from the basics of motor control and camera image acquisition to high level tasks like data collection and AI training, where you teach JetBot to follow objects, avoid collisions and more.”

1 jetson nano pressgraphic 2000x1200 Nvidia

The JetBot project for Nvidia’s Jetson Nano Dev Kit.

Sounds pretty nifty! Nvidia sent PCWorld a review unit to try out JetBot and the inferencing tutorials, but it arrived too late for us to test it before the announcement. We’re hoping to have a more extensive follow-up some point soon. If you’re interested in dabbling in deep learning projects, the Jetson Nano Developer Kit goes on sale today. Curious makers should visit the Jetson developer site for all the info you need to get started with Nvidia’s tiny, affordable AI board.

Join the newsletter!

Or

Sign up to gain exclusive access to email subscriptions, event invitations, competitions, giveaways, and much more.

Membership is free, and your security and privacy remain protected. View our privacy policy before signing up.

Error: Please check your email address.
Keep up with the latest tech news, reviews and previews by subscribing to the Good Gear Guide newsletter.
Brad Chacos

Brad Chacos

PC World (US online)
Show Comments

Most Popular Reviews

Latest Articles

Resources

PCW Evaluation Team

Jack Jeffries

MSI GS75

As the Maserati or BMW of laptops, it would fit perfectly in the hands of a professional needing firepower under the hood, sophistication and class on the surface, and gaming prowess (sports mode if you will) in between.

Taylor Carr

MSI PS63

The MSI PS63 is an amazing laptop and I would definitely consider buying one in the future.

Christopher Low

Brother RJ-4230B

This small mobile printer is exactly what I need for invoicing and other jobs such as sending fellow tradesman details or step-by-step instructions that I can easily print off from my phone or the Web.

Aysha Strobbe

Microsoft Office 365/HP Spectre x360

Microsoft Office continues to make a student’s life that little bit easier by offering reliable, easy to use, time-saving functionality, while continuing to develop new features that further enhance what is already a formidable collection of applications

Michael Hargreaves

Microsoft Office 365/Dell XPS 15 2-in-1

I’d recommend a Dell XPS 15 2-in-1 and the new Windows 10 to anyone who needs to get serious work done (before you kick back on your couch with your favourite Netflix show.)

Maryellen Rose George

Brother PT-P750W

It’s useful for office tasks as well as pragmatic labelling of equipment and storage – just don’t get too excited and label everything in sight!

Featured Content

Don’t have an account? Sign up here

Don't have an account? Sign up now

Forgot password?