Nvidia Jetson
lastmod 2026-05-12created 2022-07-2834 edits
Nvidia Jetson is a series of embedded computing boards from Nvidia. The Jetson TK1, TX1, and TX2 models all carry an Nvidia Tegra processor that includes an ARM architecture central processing unit (CPU). Jetson is a low-power system designed to accelerate machine learning applications.
| WARNING |
| If a software update dialog appears, "never" update! In my case, after updating and rebooting, LXDE showed a black screen... so I installed JetPack from scratch. |
Hardware
The Nvidia family includes the following boards:
- In late 2014, Nvidia shipped the Nvidia Jetson TK1 development board, which included a Tegra K1 SoC in the T124 model and ran Ubuntu Linux.
- The Nvidia Jetson TX1 development board carries the Tegra X1 of model T210.
- The Nvidia Jetson TX2 board carries the GP10B microarchitecture of the Tegra X2 (SoC type T186 or a similar type). This board and the related development platform were announced in March 2017 as a compact card design for low-power scenarios (e.g., use in small camera drones). A matrix describing the performance modes was provided along with it through the press. The TX2i model was mentioned as suitable for industrial use.
- The Nvidia Jetson Xavier was announced as a development kit in late August 2018. Product images and some technical specifications were released at the same time as the announcement. It was said that a 20x acceleration over the previous device was expected for certain uses, with a 10x improvement in power efficiency.
- The Nvidia Jetson Nano was announced as a development system in mid-March 2019, and pre-orders for the development kit became available in June 2019. Discussion of this product centered on hobbyist robotics, owing to its low price. The final specifications revealed it as a power-optimized, simplified version of the full Tegra X1 system.
Bluetooth support
The #Jetson AGX Xavier Developer Kit and #Jetson Nano Developer Kit could connect over Bluetooth, although it is not in the spec documents.
However, the #BOXER-8240AI-A2-1111 could not connect over Bluetooth.
My guess is that the Developer Kits include a Bluetooth module. I'll probably need to test it myself...
Installing PyTorch
ARM is not among the architectures supported on the official PyTorch site, so it has to be downloaded separately.
It also seems to depend on the JetPack version.
Jetson AGX Thor Developer Kit
Jetson AGX Thor Developer Kit
| - | Jetson T5000 | Jetson T4000* |
| AI Performance | 2070 TFLOPS (FP4—Sparse) | 1200 TFLOPS (FP4—Sparse) |
| GPU |
|
|
| CPU |
|
|
| Memory |
|
|
| Networking** | 4x 25 GbE | 3x 25 GbE |
| Storage |
| |
| Mechanical |
| |
| Power | 40 W–130 W | 40 W–70 W |
Quick Start Guide
- Go to the website and download the BSP image.
- Create a boot disk (USB) with a tool such as Etcher.
- After booting, select
Jetson Thor options->Flash Jetson Thor AGX Developer Kit on NVMe 0.2.0-r38.1and it installs all the stuff on its own.
Jetson AGX Orin Developer Kit
Jetson AGX Orin Developer Kit
The Tech Specs:
| GPU | NVIDIA Ampere architecture with 2048 NVIDIA® CUDA® cores and 64 Tensor cores |
| CPU | 12-core Arm Cortex-A78AE v8.2 64-bit CPU 3MB L2 + 6MB L3 |
| DL Accelerator | 2x NVDLA v2.0 |
| Vision Accelerator | PVA v2.0 |
| Memory | 32GB 256-bit LPDDR5 204.8 GB/s |
| Storage | 64GB eMMC 5.1 |
| Video Encode | 2x 4K60 / 4x 4K30 / 8x 1080p60 / 16x 1080p30 (H.265) |
| Video Decode | 1x 8K30 / 3x 4K60 / 6x 4K30 / 12x 1080p60 / 24x 1080p30 (H.265) |
REFERENCE CARRIER BOARD:
| Camera | 16 lane MIPI CSI-2 connector |
| PCIe | x16 PCIe slot supporting x8 PCIe Gen4 |
| RJ45 | Up to 10 GbE |
| M.2 Key M | x4 PCIe Gen 4 |
| M.2 Key E | x1 PCIe Gen 4, USB 2.0, UART, I2S |
| USB Type-C | 2x USB 3.2 Gen2 with USB-PD support |
| USB Type-A | 2x USB 3.2 Gen2, 2x USB 3.2 Gen1 |
| USB Micro-B | USB 2.0 |
| DisplayPort | DisplayPort 1.4a (+MST) |
| microSD slot | UHS-1 cards up to SDR104 mode |
| Other |
|
| Dimensions | 110mm x 110mm x 71.65mm (Height includes feet, carrier board, module, and thermal solution) |
How to install JetPack
- Jetson Orin Nano Developer Kit User Guide - Software Setup | NVIDIA Developer
- Jetson Orin Nano Developer Kit Getting Started Guide | NVIDIA Developer
- Korean blogs are no help.
Jetson-orin-nano-qtr_numbered.png
There are two methods.
Installing with SDK Manager
- [Recommended] (Jetson) Installing Jetpack on a Jetson Orin Nano SSD (using SDK Manager) — STARLANE
- Summary of the Jetson Orin nano installation process
- Download and run SDK Manager from https://developer.nvidia.com/sdk-manager. (Wait at the STEP 01 screen)
- Without plugging in the power, connect the Orin Nano's Type-C port to the host computer.
- Short the pins labeled
FC RECandGNDtogether with a female-to-female jumper cable. -> This seems to be how to enter recovery mode - In this state, connect power to the Orin Nano.
- Then proceed with the installation and so on.
Writing directly to an SD CARD
- Jetson Orin Nano Developer Kit Getting Started Guide | NVIDIA Developer
- [Recommended] 🚀 Initial Setup Guide - Jetson Orin Nano - NVIDIA Jetson AI Lab
In short, just write JetPack directly to an SD card. However, you must check the firmware version first.
- Connect a monitor to the Orin Nano and press ESC repeatedly to enter the BIOS.
- The firmware version is written on the third line from the top. Check it.
- If the version is below 36.0, first install the JetPack 5.1.3 firmware to upgrade the firmware.
- Jetson Orin Nano Developer Kit JetPack 5.1.3 image
- For how to upgrade after that, see the https://www.jetson-ai-lab.com/initial_setup_jon.html page.
- If the version is 36.x or higher, you can install JetPack 6.
For reference, when writing data to the SD card, just write the ZIP image as-is with something like Etcher.
Jetson AGX Xavier Developer Kit
Jetson AGX Xavier Developer Kit
The Tech Specs:
| GPU | 512-core Volta GPU with Tensor Cores |
| CPU | 8-core ARM v8.2 64-bit CPU, 8MB L2 + 4MB L3 |
| Memory | 32GB 256-Bit LPDDR4x / 137GB/s |
| Storage | 32GB eMMC 5.1 |
| DL Accelerator | (2x) NVDLA Engines |
| Vision Accelerator | 7-way VLIW Vision Processor |
| Encoder/Decoder | (2x) 4Kp60 / HEVC/(2x) 4Kp60 / 12-Bit Support |
| Size | 105 mm x 105 mm x 65 mm |
| Deployment | Module (Jetson AGX Xavier) |
Developer Kit I/Os:
| Developer Kit I/Os | Jetson AGX Xavier Module Interface |
| PCIe X16 | x8 PCIe Gen4/x8 SLVS-EC |
| RJ45 | Gigabit Ethernet |
| USB-C | 2x USB 3.1, DP (Optional), PD (Optional) Close-System Debug and Flashing Support on 1 Port |
| Camera Connector | (16x) CSI-2 Lanes |
| M.2 Key M | NVMe |
| M.2 Key E | PCIe x1 + USB 2.0 + UART (for Wi-Fi/LTE) / I²S / PCM |
| 40-Pin Header | UART + SPI + CAN + I²C + I²S + DMIC + GPIOs |
| HD Audio Header | High-Definition Audio |
| eSATAp + USB3.0 Type A | SATA Through PCIe x1 Bridge (PD + Data for 2.5-inch SATA) + USB 3.0 |
| HDMI Type A | HDMI 2.0 |
| uSD/UFS Card Socket | SD/UFS |
AAEON BOXER-8240AI-A2-1111
BOXER-8240AI
- AAEON BOXER-8240AI | AI@Edge Embedded BOX PC with Nvidia Jetson AGX Xavier Platform - AAEON
- BOXER-8240AI-A2-1111 | AAEON at Edge PC NVidia | SOS electronic
AI at Edge Box PC NVidia Jetson AGX Xavier 32GB RAM 32GB eMMC -10...55°C
Specifications:
| Manufacturer (Brand) | AAEON |
| Series | BOXER-8240AI |
| Processor | Nvidia Jetson AGX Xavier |
| Processor Spec. | Nvidia Carmel ARM v8.2 x8, Volta GPU 512 CUDA cores, 64 Tensor cores |
| RAM (Slot, Spec, Total) | 32GB 256-Bit LPDDR4 2133 onboard |
| Storage | eMMC 5.1 32GB, M.2 2280 M-Key (PCIe x4, SATA), 2,5" SSD SATA, microSD |
| Expansion Slots | M.2 2230 E-Key |
| Ethernet | GbE, GbE PoE 802.11af x4 |
| USB | USB 3.2 Gen 1 x3, USB 2.0 |
| Serial Com Interfaces | RS232/422/485, CAN |
| Display Interfaces | DP, HDMI |
| Concurrent Displays | 2 |
| Audio | - |
| Other Features | 40-pin header with UART, SPI, I2C, I2C, DMIC, GPIO |
| Power Supply | 12...24V DC |
| Dimensions | 210 x 164 x 59mm |
| Mounting | Wallmount |
| Operating Temperature | -10...+55°C |
| Note | Jetpack 4.4, ACLinux 4.9 |
| RoHS | Yes |
Jetson Xavier NX Developer Kit
Jetson Xavier NX Developer Kit
| GPU | NVIDIA Volta architecture with 384 NVIDIA CUDA® cores and 48 Tensor cores |
| CPU | 6-core NVIDIA Carmel ARM®v8.2 64-bit CPU 6 MB L2 + 4 MB L3 |
| DL Accelerator | 2x NVDLA Engines |
| Vision Accelerator | 7-Way VLIW Vision Processor |
| Memory | 8 GB 128-bit LPDDR4x @ 51.2GB/s |
| Storage | microSD (not included) |
| Video Encode | 2x 4K @ 30 / 6x 1080p @ 60 / 14x 1080p @ 30 (H.265/H.264) |
| Video Decode |
|
| Camera | 2x MIPI CSI-2 DPHY lanes |
| Connectivity | Gigabit Ethernet, M.2 Key E (WiFi/BT included), M.2 Key M (NVMe) |
| Display | HDMI and display port |
| USB | 4x USB 3.1, USB 2.0 Micro-B |
| Others | GPIO, I²C, I²S, SPI, UART |
| Mechanical | 103 mm x 90.5 mm x 34.66 mm |
Jetson Nano Developer Kit
Jetson Nano Developer Kit
Technical Specifications:
| GPU | 128-core Maxwell |
| CPU | Quad-core ARM A57 @ 1.43 GHz |
| Memory | 4 GB 64-bit LPDDR4 25.6 GB/s |
| Storage | microSD (not included) |
| Video Encode | 4K @ 30 / 4x 1080p @ 30 / 9x 720p @ 30 (H.264/H.265) |
| Video Decode | 4K @ 60 / 2x 4K @ 30 / 8x 1080p @ 30 / 18x 720p @ 30 (H.264/H.265) |
| Camera | 2x MIPI CSI-2 DPHY lanes |
| Connectivity | Gigabit Ethernet, M.2 Key E |
| Display | HDMI and display port |
| USB | 4x USB 3.0, USB 2.0 Micro-B |
| Others | GPIO, I²C, I²S, SPI, UART |
| Mechanical | 69 mm x 45 mm, 260-pin edge connector |
Jetson Nano 2GB Developer Kit
Jetson Nano 2GB Developer Kit
Technical Specifications:
| GPU | 128-core NVIDIA Maxwell™ |
| CPU | Quad-core ARM® A57 @ 1.43 GHz |
| Memory | 2 GB 64-bit LPDDR4 25.6 GB/s |
| Storage | microSD (Card not included) |
| Video Encode | 4Kp30 / 4x 1080p30 / 9x 720p30 (H.264/H.265) |
| Video Decode | 4Kp60 / 2x 4Kp30 / 8x 1080p30 / 18x 720p30 (H.264/H.265) |
| Connectivity | Gigabit Ethernet, 802.11ac wireless1 |
| Camera | 1x MIPI CSI-2 connector |
| Display | HDMI |
| USB | 1x USB 3.0 Type A,2x USB 2.0 Type A, USB 2.0 Micro-B |
| Others |
|
| Mechanical | 100 mm x 80 mm x 29 mm |
Changing the operating mode
- (Jetson) (nvpmodel) Performance and fan mode settings
- JETSON TX2 performance and changing the operating mode
Two executables are used: nvpmodel and jetson_clocks.
Maximum performance mode:
Market trends and major manufacturers as of 2026
The NVIDIA Jetson series holds an overwhelming share (about 39% or more) of the industrial edge computing market. The sales ranking is usually determined by price accessibility, performance (TOPS), and performance per watt.
Below is a summary of the sales popularity ranking and recommended products, based on current 2026 market trends and shipments from major manufacturers (ASUS, Advantech, AAEON, etc.).
- #1: Jetson Orin Nano series (entry level)
- The lineup with the highest sales. Thanks to its low price and compact size, it is the most common choice in settings that need mass deployment, such as smart cameras, small robots, and drones.
- Key features: 7W–15W low power, up to 40–67 TOPS of AI performance.
- Recommended products:
- ASUS PE1100N / PE1101N: Stable enough to be called the industrial standard, and highly versatile with support for a variety of I/O.
- Advantech MIC-711-OX: Sells widely for factory automation lines thanks to its reliable thermal design.
- Seeed Studio reComputer J4012: Easy for developers to access, so it is popular at the stage of moving from prototype to mass production.
- The model with the best balance between performance and size. It keeps the same form factor as the Orin Nano but is much more powerful (up to 100 TOPS), so it mainly sells for multi-camera analytics and high-performance autonomous mobile robots (AMR).
- Key features: 10W–25W power, about 3–5x the performance of the Orin Nano.
- Recommended products:
- ASUS PE1102N: Supports the GMSL2 camera interface, making it very popular with autonomous vehicle/robot development teams.
- Neousys NRU-52S / 51V: A bestseller in outdoor mobility thanks to a design that withstands harsh environments (vibration, temperature).
- Bought by customers who want server-class performance at the edge. It is installed in smart city traffic control centers and equipment that must run complex multimodal AI models.
- Key features: 15W–60W power, overwhelming performance of up to 275 TOPS.
- Recommended products:
- ASUS PE2100N: Supports 10G LAN and PoE ports, specialized for security monitoring systems that connect many high-resolution IP cameras.
- ADLINK DLAP-411-Orin: High sales in high-end industries such as medical devices and precision inspection equipment.
New products to watch in 2026
The recently released Jetson AGX Thor is based on the Blackwell architecture and delivers more than 2000 TOPS.
It is currently in pre-sale and early adoption, centered on humanoid robots and cutting-edge autonomous driving, and is expected to become the mainstream of the high-end market.
Selection guide summary:
| Use case | Recommended lineup | Representative model |
| Cost-effectiveness & mass deployment | Orin Nano | ASUS PE1100N |
| Robots & intelligent video analytics | Orin NX | "ASUS PE1102N, Neousys NRU series" |
| Top performance & smart city | AGX Orin | ASUS PE2100N |
See also
Favorite site
References
-
Not initially available in all regions ↩