Nvidia has unveiled the Jetson Nano, an “AI computer that makes it possible to create millions of intelligent systems”.
The module delivers 472 GFLOPS of compute performance and can consume as little as 5W of power, said Nvidia.
The developer kit version of the Jetson Nano is aimed at developers, makers, and enthusiasts – and supports high-resolution sensors and popular AI frameworks.
“Jetson Nano makes AI more accessible to everyone and is supported by the same underlying architecture and software that powers our nation’s supercomputers,” said Nvidia.
The developer kit comes with out-of-the-box support for full desktop Linux, compatibility with many peripherals and accessories, and ready-to-use projects and tutorials that help makers get started. A Jetson developer forum is also available to users.
Nvidia said the key hardware features of the Jetson Nano are its 128-core Nvidia Maxwell architecture-based GPU and Quad-core ARM A57 CPU.
This has been combined with 4GB of RAM and 16GB of flash memory.
The Jetson Nano developer kit is available for pre-order from Nvidia for $99, while the production-ready Jetson Nano module for companies will cost $129.
It is set to ship worldwide in June 2019.
|Jetson Nano Module|
|CPU||Quad-core ARM Cortex-A57 MPCore|
|GPU||Maxwell Architecture with 128 Nvidia CUDA cores|
|Storage||16GB eMMC 5.1 Flash|
|Display||HDMI 2.0 or DP1.2 | eDP 1.4 | DSI (1 x2)|
|Camera||12 lanes MIPI CSI-2 DPHY 1.1 (1.5Gbps)|
|I/O||1x SDIO / 2x SPI / 6x I2C / 2x I2S / GPIOs|
|Video Encode/Decode||4K @ 30Hz/60Hz|
|Dimensions||69.6 x 45 mm|
In addition to the module specifications above, the Jetson Nano developer kit will feature the following I/O:
- 4x USB 3.0, USB 2.0 Micro-B
- 1x MIPI CSI-2 DPHY lanes
- MicroSD slot
- HDMI 2.0 and eDP 1.4
- GPIO, I2C, I2S, SPI, UART