
1 / 4 · Click to enlarge
AI Summary
A compact edge AI box with NVIDIA Jetson Orin Nano 8GB, 40 TOPS performance, and pre-installed JetPack on a 128GB NVMe SSD. It offers rich I/O including 4x USB 3.2, HDMI, GbE, and M.2 expansion for rapid AI deployment.
Key Highlights
- NVIDIA Jetson Orin Nano 8GB module with 40 TOPS AI performance
- 1024-core Ampere GPU with 32 Tensor Cores for accelerated inference
- 6-core Arm Cortex-A78AE CPU running up to 1.5 GHz
- Pre-installed JetPack on 128GB NVMe SSD for immediate use
- 4x USB 3.2, HDMI, GbE, M.2 Key E and Key M slots
- Compact aluminum case with 7W to 15W power rating
Ideal For
- AI developers building edge inference prototypes
- System integrators deploying video analytics
- Robotics engineers needing compact AI compute
- Smart city and security solution providers
- Industrial automation teams implementing vision systems
Buying Tip
Remember to order a compatible power adapter separately, as none is included, and confirm the input voltage required by the carrier board before first use.
Elecvora expert review
Our verdict
The reComputer J3011 offers a well-integrated Jetson Orin Nano 8GB platform with pre-installed JetPack and a 128GB SSD, reducing setup time for edge AI projects. Its rich I/O and compact metal case make it a solid choice for production deployments where reliability and expandability matter.
AI-assisted analysis based on the manufacturer's specifications. Not a hands-on lab test.
Pros
- Powerful 40 TOPS AI performance in a small form factor
- Pre-installed JetPack on NVMe SSD speeds up development
- Extensive connectivity including 4x USB 3.2 and dual M.2 slots
- Aluminum case with mounting options for flexible installation
Cons
- Power adapter not included, requiring a separate purchase
- No built-in WiFi/BT until the included module is installed
- Video encoding relies on CPU cores, limiting encode performance
Best for
AI developers and integrators needing a ready-to-use Jetson Orin Nano edge device
Not ideal for
Users who require a complete plug-and-play kit with power supply
Specifications
| RAM | 8GB |
|---|---|
| Module | Jetson Orin Nano 8GB |
| AI Performance | 40 TOPS |
| GPU | 1,024-core NVIDIA Ampere architecture GPU with 32 Tensor Cores |
| Max GPU Frequency(MHz) | 625 MHz |
| CPU | 6-core Arm® Cortex®-A78AE v8.2 64-bit CPU 1.5MB L2 + 4MB L3 |
| Max CPU Frequency(MHz) | 1.5 GHz |
| Storage Capacity | (Supports external NVMe) |
| Video encoder | 1,080p30 supported by 1-2 CPU cores |
| Video decoder | 1x 4K60 (H.265), 2x 4K30 (H.265), 5x 1,080p60 (H.265), 11x 1,080p30 (H.265) |
| Camera Details | Up to 4 cameras (8 via virtual channels***) 8 lanes MIPI CSI-2 D-PHY 2.1 (up to 20Gbps) |
| PCIE | 1 x4 + 3 x1 (PCIe Gen3, Root Port, & Endpoint) |
| Power Rating | 7W - 15W |
| Other I/O | 3x UART, 2x SPI, 2x I2S, 4x I2C, 1x CAN, DMIC & DSPK, PWM, GPIOs |
| Display | 1x 4K30 multi-mode DP 1.2 (+MST)/eDP 1.4/HDMI 1.4** |
| Networking | 1x GbE |
| Mechanical | 69.6 mm x 45 mm 260-pin SO-DIMM connector |
| Shipping Weight | 0.2 kg |
| Shipping Dimensions | 12 × 10 × 7 cm |
What's in the box
- reComputer J3011 device with Jetson Orin Nano 8GB module
- Heatsink
- 128GB NVMe SSD with JetPack pre-installed
- WiFi/BT combo module
- 2x antennas
Frequently Asked Questions
What operating system comes pre-installed?
Do I need to buy a power adapter separately?
Can I add more storage?
How many cameras can I connect?
Is WiFi and Bluetooth included?
What is the power consumption?
Overview
The Seeed Studio reComputer J3011 is a ready-to-deploy edge AI device powered by the NVIDIA Jetson Orin Nano 8GB module. It packs up to 40 TOPS of AI performance into a small aluminum chassis, making it suitable for on-device processing in space-constrained installations. The system includes a production-grade Jetson Orin Nano module, a heatsink, and a carrier board with extensive connectivity. A 128GB NVMe SSD comes pre-installed with JetPack, so you can begin development without assembling storage or flashing an OS manually. Whether you are prototyping a vision system or rolling out edge inference across a fleet, the J3011 provides a consistent hardware foundation.
Key Features
- High AI throughput: The 1024-core NVIDIA Ampere GPU with 32 Tensor Cores delivers up to 40 TOPS, enabling real-time inference for vision and NLP models.
- Efficient CPU: A 6-core Arm Cortex-A78AE v8.2 64-bit CPU with 1.5MB L2 and 4MB L3 cache handles control tasks and data preprocessing.
- Flexible storage: The included 128GB NVMe SSD provides fast local storage; an M.2 Key M slot supports additional NVMe drives.
- Rich I/O: Four USB 3.2 ports, HDMI, Gigabit Ethernet, M.2 Key E for WiFi/BT, M.2 Key M, CAN, GPIO, I2C, I2S, UART, SPI, and MIPI CSI-2 camera interfaces.
- Compact and mountable: The aluminum case measures approximately 130 x 120 x 58.5 mm and supports desktop or wall mounting.
- Power efficient: Operates from 7W to 15W, suitable for fanless or low-power edge deployments.
Specifications Explained
The Jetson Orin Nano 8GB module is the heart of the system. Its GPU architecture is designed for parallel processing, which accelerates neural network inference. The 40 TOPS figure indicates the maximum AI compute available, useful for tasks like object detection and semantic segmentation. The 8GB of RAM is shared between CPU and GPU, so model size and batch size should be planned accordingly. Video encode is handled by 1-2 CPU cores at 1080p30, while the dedicated decoder supports up to 1x 4K60, 2x 4K30, or multiple 1080p streams, making the device capable of handling several camera feeds. The MIPI CSI-2 interface supports up to four physical cameras (eight via virtual channels), with 8 lanes at up to 20Gbps total. For expansion, the M.2 Key E slot accepts a WiFi/BT module (a combo module and two antennas are included), and the M.2 Key M slot is available for additional NVMe storage. The 260-pin SO-DIMM connector is the standard interface for the Jetson module, but the module is already installed in the carrier board.
Compatibility and Requirements
The reComputer J3011 ships with JetPack pre-installed on the 128GB NVMe SSD, providing a Linux environment and support for leading AI frameworks. A power adapter is not included; you will need a suitable power supply within the 7W to 15W range. The device supports desktop and wall mounting, but mounting hardware is not listed. For camera integration, MIPI CSI-2 cameras are required; USB cameras can also be used via the USB 3.2 ports. The M.2 Key E slot is intended for a WiFi/BT module, and a combo module with antennas is included. Ensure any additional NVMe drive for the M.2 Key M slot is compatible with the PCIe Gen3 interface.
Typical Uses
This edge AI device is well suited to video analytics, object detection, pose estimation, semantic segmentation, natural language processing, medical imaging, and robotics. It can be deployed in smart cities, security systems, industrial automation, and smart factories. The compact size and low power envelope make it practical for embedding into autonomous machines or remote monitoring stations.
What's in the Box
The package includes the reComputer J3011 device with the Jetson Orin Nano 8GB module installed, a heatsink, a 128GB NVMe SSD with JetPack pre-installed, a WiFi/BT combo module, and two antennas. A power adapter is not included.

Quick specs
- RAM
- 8GB
- Module
- Jetson Orin Nano 8GB
- AI Performance
- 40 TOPS
- GPU
- 1,024-core NVIDIA Ampere architecture GPU with 32 Tensor Cores
- Max GPU Frequency(MHz)
- 625 MHz
- CPU
- 6-core Arm® Cortex®-A78AE v8.2 64-bit CPU 1.5MB L2 + 4MB L3



















