Product description: reComputer Mini J4012 with expansion board - Nvidia Orin NX 16 GB RAM + 128 GB NVMe - Seeedstudio 114993553
reComputer Mini J4012 with Seeedstudio extension is a modern solution for professionals and enthusiasts of AI technology. Thanks to the use of an efficient Nvidia Jetson Orin NX module with 16 GB of RAM , the device provides up to 100 TOPS of power to support the most advanced artificial intelligence algorithms and image analysis. The compact design with dimensions of 63 x 95 x 66.7 mm and a weight of 462 g means that the reComputer can be easily mounted on both a desk and a wall, which increases implementation versatility. The built-in PCIe connector allows for expansion with additional modules, and a wide range of interfaces, including 8 USB ports, CAN, Ethernet, DisplayPort , allows for integration in various applications. Preinstalled JetPack 6.0 on a fast 128 GB NVMe SSD drive and support for OTA remote management by Allxon / Balena speed up the implementation process and facilitate the scaling of projects.
Key Features of the reComputer Mini J4012 with Expansion Board
- AI performance up to 100 TOPS : Enables the most demanding deep learning and vision AI algorithms
- Jetson Orin NX 16GB Module : Based on Nvidia Ampere GPU with 1024 cores and 32 Tensor Cores
- Multi-channel video processing : supports up to 18 1080p 30fps streams simultaneously
- Rich set of interfaces : up to 8x USB, 2x CAN, 1x Ethernet, 1x DisplayPort, M.2 Key E, M.2 Key M
- Compact housing : comfortable dimensions of 63 x 95 x 66.7 mm and weight of 462 g
- Resistance to harsh conditions : supports voltages from 12 V to 54 V and operates in temperatures from -10°C to 50°C
- Ready-made development environment : JetPack 6.0 pre-installed with support for TensorRT, CUDA, cuDNN and DeepStream
- Remote management and scaling : support for Allxon and BalenaCloud
Minicomputer interface layout
The design of the reComputer Mini J4012 has been optimized for compactness and convenient access to all ports and interfaces. The mini computer is equipped with a cooling fan, communication connectors and expansion ports located on the front and rear panels.
Elements arranged on the housing:
- Reset button
- HOST+DP connector (USB Type-C with DisplayPort 1.4 support)
- USB 2.0 microB port (Device Mode)
- Recovery button
- Power connector (12V to 54V DC, XT30)
- USB 2.0 port
- UART port 1
- UART DEBUG port
Components on the lower expansion board:
- 2x USB 3.2 Type-A (10Gb/s)
- 4x USB 3.0 Type-A (5Gb/s)
- CAN0 + Power connector (power supply and CAN bus)
- IIC1 (I2C) interface
- IIC0 (I2C) interface
- Can1 interface (CAN bus)
- RJ45 Ethernet port (Gigabit Ethernet)
- SPI port (serial peripheral communication)
reComputer Mini J4012 Software
The reComputer Mini J4012 comes pre-loaded with JetPack 6.0 based on Linux Ubuntu, installed on a fast 128 GB NVMe SSD . It supports key libraries and frameworks such as TensorRT, CUDA, cuDNN, DeepStream, TensorFlow and PyTorch . Thanks to the prepared jetson-examples projects, it is possible to quickly implement generative AI solutions (Ollama, Llama3) and video analysis (YOLOv8). Remote management and OTA updates are implemented using the Allxon and BalenaCloud platforms.
Practical applications
The reComputer Mini J4012 with the expansion is a versatile solution for edge AI projects that require high performance at low power consumption. It is ideal for building autonomous patrol robots, delivery drones and intelligent industrial IoT systems. With the ability to support up to 18 1080p video streams at 30 fps, the computer is suitable for advanced monitoring and video analytics systems using Vision-Language Models (VLM) and generative AI technologies. The reComputer Mini is also used to implement applications for image recognition (e.g. YOLOv8), video data analysis and local content processing in the industrial, transport and security sectors.
Kit Contents
- 1x reComputer Mini J4012 with expansion
- 2x WiFi Antenna
- 1x XT30-DC Power Cable
| Technical specification of reComputer Mini J4012 | |
|---|---|
| Processor | Nvidia Jetson Orin NX 16GB |
| AI Performance | 100 TOPS |
| GPU | 1024-core Nvidia Ampere architecture GPU with 32 Tensor Cores |
| CPU | 8-core Arm Cortex-A78AE v8.2 64-bit |
| RAM memory | 16GB 128bit LPDDR5 |
| Mass storage | 128GB NVMe SSD |
| Video encoder | 1x 8K30 (H.265), 2x 4K60 (H.265), 4x 4K30 (H.265), 9x 1080p60 (H.265), 18x 1080p30 (H.265) |
| Video decoder | 1x 8K30 (H.265), 2x 4K60 (H.265), 4x 4K30 (H.265), 9x 1080p60 (H.265), 18x 1080p30 (H.265) |
| DL Accelerator | 2x NVDLA v2 |
| USB interfaces | 2x USB 3.2 Type-A (10 Gb/s), 1x USB 2.0 microB (Device mode), 1x USB 3.2 Type-C (Host mode), 1x USB 2.0 JST 5-pin (Host mode) |
| Ethernet | 1x RJ45 Gigabit Ethernet (10/100/1000M) |
| CAN | 2x CAN (JST 4-pin and XT30 2+2) |
| I2C | 2x I2C JST 4-pin (3.3V) |
| SPI | 1x SPI JST 6-pin (3.3V) |
| M.2 Connectors | M.2 Key E (WiFi/Bluetooth), M.2 Key M (NVMe SSD) |
| RTC | 1x RTC 2-pin |
| Power supply | 1x XT30 (12V to 54V DC) |
| Fan | 1x 4-pin PWM (5V) |
| Operating temperature | from -10°C to 50°C |
| Dimensions | 63 x 95 x 66.7 mm |
| Mass | 462g |




