Details
Advanced AI computer for smaller, lower-power autonomous machines
Jetson Orin NX 16GB module delivers up to 100 TOPS of AI performance in a compact and small form factor, with power configurable between 10W and 25W, up to 3X the performance of Jetson AGX Xavier and 5X the performance of Jetson Xavier NX. Upgrading to the SUPER version delivers up to 157 TOPS with power options between 10W and 40W.
The Jetson Orin NX module can be upgraded to SUPER Mode via a software upgrade, improving performance by 70%. We provide upgrade tutorial in the wiki to help users upgrade quickly
| Version | 8GB*** SUPER |
16GB*** SUPER |
|---|---|---|
| AI performance | 117 TOPS | 157 TOPS |
| GPU | 1024-core N-VIDIA Ampere architecture GPU with 32 Tensor Cores | |
| GPU Frequency | 1173MHz (Max) | |
| CPU | 6-core Arm® Cortex®-A78AE v8.2 64-bit CPU, 2MB L2 + 4MB L3 | 8-core Arm® Cortex®-A78AE v8.2 64-bit CPU, 2MB L2 + 4MB L3 |
| CPU Frequency | 2GHz (Max) | |
| DL accelerator | 1x N-VDLA v2 | 2x N-VDLA v2 |
| DLA Frequency | 1230MHz (Max) | |
| Vision Accelerator | 1x PVA v2 | |
| Memory | 8GB 128-bit LPDDR5, 102.4GB/s | 16GB 128-bit LPDDR5, 102.4GB/s |
| Storage | Supports external NVMe | |
| Power | 10W | 15W | 25W | 40W | |
| PCIe* | 1 x4 + 3 x1 (PCIe Gen4, Root Port, & Endpoint) | |
| USB* | 3x USB 3.2 2.0 (10 Gbps) 3x USB 2.0 |
|
| CSI Camera | Up to 4 cameras (8 via virtual channels**) 8 lanes MIPI CSI-2 D-PHY 1.2 (up to 20 Gbps) |
|
| Video encode | 1x 4K60 (H.265) | 3x 4K30 (H.265) 6x 1080p60 (H.265) | 12x 1080p30 (H.265) |
|
| Video decode | 1x 8K30 (H.265) | 2x 4K60 (H.265) | 4x 4K30 (H.265) 9x 1080p60 (H.265) | 18x 1080p30 (H.265) |
|
| Display | 1x 8K30 multi-mode DP 1.4a (+MST)/eDP 1.4a/HDMI 2.1 | |
| Other | 3x UART, 2x SPI, 2x I2S, 4x I2C, 1x CAN, DMIC & DSPK, PWM, GPIO | |
| Networking | 1x GbE | |
| Dimensions | 69.6 mm x 45 mm 260-pin SO-DIMM connector |
|
| * USB 3.2, MGBE, and PCIe share UPHY Lanes. See the Product Design Guide for supported UPHY configurations. ** Virtual Channels for Jetson Orin NX are subject to change Refer to the Software Features section of the latest NVIDIA Jetson Linux Developer Guide for a list of supported features. *** Orin NX Module and Waveshare kit: supports upgrading to SUPER image, provides SUPER upgrade tutorial in the wiki (jetpack 6.2 can support upgrading SUPER version); |
||
N-VIDIA Software
All N-VIDIA Jetson modules and developer kits are supported by the same N-VIDIA Jetson software stack, so you can develop once and deploy everywhere. Jetson software is designed to provide end-to-end acceleration for AI applications and accelerate the time to market. This brings the powerful N-VIDIA technologies that power data centers and cloud deployments to the edge.
Pretrained Models
Many AI applications have common needs: classification, object detection, language translation, text-to-speech, recommender engines, sentiment analysis, and more. Pretrained models from the NGC™ catalog are performance-optimized and ready to be fine-tuned with the N-VIDIA TAO Toolkit and customer datasets, reducing time and cost for production AI software development.
N-VIDIA software platforms including Isaac for robotics, Riva for AI speech, and Metropolis for video analytics, are available to speed vertical-specific applications
N-VIDIA Isaac
Robots improve efficiency and quality of life in industries such as manufacturing, logistics, healthcare, and service. N-VIDIA Isaac™ accelerates the process with enhanced robotics development, simulation, and deployment.
N-VIDIA Riva
N-VIDIA Riva is a fully accelerated SDK for building multimodal conversational AI applications using an end-to-end deep-learning pipeline, suitable for optimized skills in speech, vision, and natural language processing (NLP) tasks.
N-VIDIA Metropolis
N-VIDIA Metropolis includes an application framework, s set of developer tools, and the partner ecosystem. It brings visual data and AI together to improve operational efficiency and safety across a broad range of industries.
| Model | Module | Developer Kit |
|---|---|---|
| Jetson Nano N-VIDIA Maxwell architecture 128 N-VIDIA CUDA cores |
 69.6×45mm 260-pin edge connector |
 100×80×29mm Official Kit: 4GB / 2GB |
 100×80×29mm Waveshare Kit |
||
| Jetson TX2 N-VIDIA Pascal architecture 256 N-VIDIA CUDA cores |
 87×50mm 400-pin connector with Thermal Transfer Plate |
 170×170×15.6mm Official Kit |
| Jetson TX2 NX N-VIDIA Pascal architecture 256 N-VIDIA CUDA cores |
 69.6×45mm 260-pin edge connector |
 103×90.5×34mm Waveshare Kit |
| Jetson Xavier NX N-VIDIA Volta architecture 384 N-VIDIA CUDA cores 48 Tensor cores |
 69.5×45mm 260-pin edge connector |
 103×90.5×31mm Official Kit |
 103×90.5×34mm Waveshare Kit: 8GB / 16GB |
||
| Jetson AGX Xavier N-VIDIA Volta architecture 512 N-VIDIA CUDA cores 64 Tensor cores |
 100×87mm 699-pin connector Integrated Thermal Transfer Plate |
 105×105×65mm |
| Jetson Orin Nano N-VIDIA Ampere architecture 512/1024 N-VIDIA CUDA cores 16/32 Tensor cores |
 Memory: 4GB / 8GB 69.6×45mm 260-pin SO-DIMM connector |
 100×79×21mm Official Kit |
 100×79×21mm Official Super Kit |
||
103×90.5×34mm Waveshare Kit: 4GB / 8GB |
||
| Jetson Orin NX N-VIDIA Ampere architecture 1024 N-VIDIA CUDA cores 32 Tensor cores |
 Memory: 8GB / 16GB 69.6×45mm 260-pin SO-DIMM connector |
103×90.5×34mm Waveshare Kit: 8GB / 16GB |
| Note: some models haven't been released yet, please contact the Waveshare service team if you need them. | ||
As an NVIDIA official authorized partner, Waveshare is capable of providing genuine products and reliable tech support
Quick Overview
OPTIONS Jetson Orin NX x1
