The Jetson TX2 NX is designed to simplify the development and deployment of Edge AI products. Thanks to the NVIDIA JetPack SDK, it supports the main AI frameworks as well as accelerated libraries for computer vision, multimedia, graphics, and embedded inference. NVIDIA also highlights compatibility with NVIDIA NGC pretrained models and tools such as the TAO Toolkit, helping reduce time to market and simplify software updates.
Technical specifications of the NVIDIA Jetson TX2 NX module
- GPU architecture: NVIDIA Pascal
- GPU: 256 NVIDIA CUDA cores
- AI performance: up to 1.33 TFLOPS
- Max GPU frequency: up to 1.3 GHz
- CPU: dual-core NVIDIA Denver 2 64-bit + quad-core Arm Cortex-A57 MPCore
- Max CPU frequency: up to 2.0 GHz
- Memory: 4 GB LPDDR4 128-bit
- Memory bandwidth: 51.2 GB/s
- Storage: 16 GB eMMC 5.1
- Power consumption: 7.5 W to 15 W
- Power supply: 5 V
- Network: 10/100/1000 Base-T Ethernet
- Cameras: up to 5 cameras (up to 12 via virtual channels)
- Camera interface: 12 MIPI CSI-2 lanes (3x4 or 5x2), D-PHY 1.2 up to 30 Gbps
- H.265 video encoding: 1x 4Kp60 / 3x 4Kp30 / 4x 1080p60 / 8x 1080p30
- H.264 video encoding: 1x 4Kp60 / 3x 4Kp30 / 7x 1080p60 / 14x 1080p30
- Video decoding: 2x 4Kp60 / 4x 4Kp30 / 7x 1080p60 / 14x 1080p30 (H.265 & H.264)
- Display: 2x multi-mode DP 1.2 / eDP 1.4 / HDMI 2.0, 2x DSI (1.5 Gbps/lane)
- Other I/O: 1x SDIO, 2x SPI, 3x UART, 4x I2S, 4x I2C, 1x CAN, GPIOs
- Mechanical form factor: 45 mm x 69.6 mm
- Connector: 260-pin SO-DIMM edge connector
- Mechanical compatibility: form factor and pin compatibility with Jetson Nano and Jetson Xavier NX
- Carrier board compatibility: compatible with the Jetson Xavier NX developer kit carrier board
Resources for the NVIDIA Jetson TX2 NX module