Jetson Nano Brings AI Computing to Everyone

Originally published at: https://developer.nvidia.com/blog/jetson-nano-ai-computing/

Looking for more? Check out the hands-on DLI training course: Getting Started with AI on Jetson Nano Update: Jetson Nano and JetBot webinars. We’ve received a high level of interest in Jetson Nano and JetBot, so we’re hosting two webinars to cover these topics. The Jetson Nano webinar discusses how to implement machine learning frameworks, develop…

Nice! I have several applications in mind already. Can't find it on any of the sales channels mentioned though :( When will the DevKit actually be available?

Hi Adam, the distributors will be taking their webpages online shortly. The NVIDIA webstore is currently taking orders and will ship out next week or before: http://store.nvidia.com/sto...

Excellent~ Can't wait to have one~ I am from Taiwan and just wrote a blog about Jetson nano, pls check http://blog.cavedu.com/2019...

I just bought a Jetson Nano from the Nvidia store. Looks like fun! Can't wait to get my hands on it!

Question. Can you use 64GB micro SD cards with the Jetson Nano? The reason I ask: I tried using a 64GB card with a Raspberry Pi. The Windows formatting program wrote an exFAT filesystem on a 64GB card, and the Raspberry Pi couldn't understand it. I had to use a different formatting program, which would write a FAT32 filesystem on a 64GB card. Then, the Raspberry Pi NOOB OS installer worked.

That's great, thanks for posting your blog!

Hi Jeff, yes you can use a 64GB microSD card with the Nano. I have used several 64GB cards with it. The image that's written to it (with Etcher tool or similar) will automatically be in Linux EXT4 format.

Hi Dustin when and where can i buy a Jetbot car kit including all the accessories? And how difficult is it to assemble it? Will high school students be able to assemble it?Thank you very much!!!

Its out of stock. Any idea about new stocks?

Can I use ssd instead of SD card? I saw m.2 port there.

Hi xsebax, the M.2 port is an M.2 Key-E, so it is for wireless cards (WLAN/BT). Booting is supported via the microSD card slot (16GB UHS-1 recommended minimum)

Hi Hannah, you can find the JetBot parts list to order here: https://github.com/NVIDIA-A...

The chassis is 3D printed, which can be printed on an Ultimaker, MakerBot, Monoprice Voxel, or other common 3D printer. Or you can use a 3D printing service that will print the parts and ship them to you. See https://github.com/NVIDIA-A... for more info about the 3D printing.

The JetBot kit is simple to assemble and should be possible for students. The hardware assembly instructions are documented here: https://github.com/NVIDIA-A...

Thank you for answer. It's a pity that they didn't think about hdd connection.

Thank you Dustin!! Your answer is very helpful!!
One more question though, does the JetBot come with any ready-to-use project code for class demo purpose?

At the moment we are using Intel NUC computers for our standalone usage. We use some java script, OpenCV C++, Ubuntu for our purpose. What can be said about Jetson Nano for our usage? Is it possible to use Jetson Nano for out setup? Can it be used for some development work also? Will it stand in-front of depth image, point cloud processing with OpenCV or PCL? If yes, what would be the ideal combination of peripherals and software for Jetson Nano?

does nvidia-docker work with this?

Hi @disqus_xdWaNNFiI4:disqus, nvidia-docker is not supported on Tegra, but upstream Docker does work on the Jetson's.

See this example script for enabling GPU passthrough: https://gist.github.com/Jas...

Hi @prasannaroutray:disqus, yes you can run these libraries and programming languages on Jetson Nano as well. You can also install ROS Melodic on Nano if you are interested in that too. Very similar to the x86/x64 distribution of Ubuntu 18.04, the aarch64 version of Ubuntu 18.04 includes many thousands of pre-built packages in the apt repository.

Hi @aniketphatak:disqus , we have updated the Jetson Nano distributors that are available online here: https://devtalk.nvidia.com/...

They will be shipping next week.

On the post it is listed that the Jetson Nano is able to run tiny-yolo at 25fps and tensorflow SSD at 27fps. Do you have instructions/sample code on how to run that on a Jetson Nano, as well as a Jetson TX2/Xavier?