Overview
reComputer Classic J3011 is an Edge AI Computer built on the NVIDIA Jetson Orin Nano 8GB module. This compact system (130mm x 120mm x 58.5mm) ships with JetPack 5.1.1 pre-installed on a 128GB NVMe SSD (original J3011 with JP5.1.3; Super Mode available via JP6.2 upgrade). It provides rich I/O including 4x USB 3.2 Type-A (10Gbps), USB2.0 Type-C (Device Mode), HDMI 2.1, RJ-45 Gigabit Ethernet, 2x CSI camera ports, CAN, RTC, and 40-pin expansion header. This SKU is w/o power adapter.
Key Features
- Jetson Orin Nano 8GB SoM with NVIDIA Ampere architecture GPU (1,024 CUDA cores, 32 Tensor Cores).
- AI performance: 40 TOPS INT8 (Sparse) / 20 TOPS (Dense) in original mode; upgrade to Super Mode (JP6.2) for up to 67 TOPS (Sparse) / 33 TOPS (Dense).
- CPU: 6-core Arm Cortex-A78AE v8.2 64-bit.
- Memory: 8GB 128-bit LPDDR5.
- Pre-installed JetPack 5.1.1 on 128GB M.2 NVMe 2280 SSD; Wi-Fi/Bluetooth combo module pre-installed via M.2 Key E.
- I/O: 4x USB 3.2 Type-A (10Gbps), 1x USB2.0 Type-C (Device Mode), 1x HDMI 2.1, 2x CSI (2-lane 15pin), 1x RJ-45 GbE (10/100/1000M), 1x CAN, 4-pin Fan connector (5V PWM), 40-pin expansion header, 12-pin control/UART header, RTC 2-pin (supports CR1220, not included).
- Power input: 919V DC; DC jack 12V/5A (adapter not included).
- Operating temperature: -10 to 60; desk and wall mounting supported.
- Remote management and OTA support via Allxon and Balena; flexible customization of accessories, logo, and interfaces based on the J3011 design.
Specifications
| System | Edge AI Computer with NVIDIA Jetson Orin Nano 8GB |
| GPU | 1024-core NVIDIA Ampere architecture GPU with 32 Tensor Cores |
| CPU | 6-core Arm Cortex-A78AE v8.2 64-bit; 1.5MB L2 + 4MB L3 |
| Memory | 8GB 128-bit LPDDR5 |
| Storage | 1x M.2 Key M PCIe (M.2 NVMe 2280 SSD 128G included) |
| Networking | 1x RJ-45 Gigabit Ethernet (10/100/1000M) |
| Wireless | 1x M.2 Key E (pre-installed Wi-Fi/Bluetooth combo module) |
| USB | 4x USB 3.2 Type-A (10Gbps), 1x USB2.0 Type-C (Device Mode) |
| Camera | 2x CSI (2-lane 15pin) |
| Display | 1x HDMI 2.1 |
| Fan | 1x 4-pin Fan Connector (5V PWM) |
| CAN | 1x CAN |
| Expansion | 1x 40-pin expansion header; 1x 12-pin control and UART header |
| RTC | 1x RTC 2-pin (supports CR1220, not included) |
| Video Encoder | 1080p30 supported by 12 CPU cores |
| Video Decoder | 1x 4K60 (H.265) | 2x 4K30 (H.265) | 5x 1080p60 (H.265) | 11x 1080p30 (H.265) |
| Power Input | 919V DC; DC jack 12V/5A |
| Dimensions (W x D x H) | 130mm x 120mm x 58.5mm (with case) |
| Installation | Desk, wall-mounting |
| Operating Temperature | -10 ~ 60 |
| Warranty | 1 Year |
Performance (Original vs Super Mode JP6.2)
| GPU Clock | Original: 635 MHz | Super Mode: 1,020MHz |
| AI Performance | Original: 40 INT8 TOPS (Sparse) / 20 INT8 TOPS (Dense) | Super Mode: 67 TOPS (Sparse) / 33 TOPS (Dense) |
| CPU Frequency | Original: 1.5 GHz | Super Mode: 1.7 GHz |
| Memory Bandwidth | Original: 68 GB/s | Super Mode: 102 GB/s |
| Module Power | Original: 7W | 15W | Super Mode: 10W | 15W | 25W |
Whats Included
- reComputer J3011 (System Installed) x1
Applications
- AI Video Analytics
- Machine Vision
- Autonomous Mobile Robot (AMR)
- Generative AI
Capable of Bringing Generative AI to the Edge
Build AI agents with Vision-Language Models (VLM) such as LLaVA to process live or archived video and images, enabling natural-language summarization, search, and insight extraction.
Build Multi-stream AI Video Analytics
Jetson Orin Nano integrates the NVIDIA Multi-Standard Video Decoder. reComputer J3011 can take 11x1080p30 streams. See the YOLOv8 DeepStream multi-stream benchmarks.
Fastest Way to Deploy Generative AI and Computer Vision Models
Use the prepared jetson-example repository for one-line deployment of edge AI projects including Ollama, Llama3, LLaVA, YOLOv8, and more.
Notes
- This J3011 variant is shipped without a power adapter. If you need the version with power adapter, please see reComputer J3011 with power adapter.
- You can self-upgrade to JetPack 6.2 to enable Super Mode and boost performance from 40 TOPS to 67 TOPS (Sparse), comparable to the NVIDIA Jetson Orin Nano Super Developer Kit.
- For optimal performance, Seeed recommends 128GB / 256GB / 512GB / 1TB NVMe SSDs from Seeed. Some market SSDs may only be compatible with specific JetPack versions, potentially affecting device functionality.
Manuals
DatasheetSchematic3D FileSeeed Nvidia Jetson Product CatalogNvidia Jetson ComparisonNvidia Jetson Product ComparisonSeeed Nvidia Jetson Success CasesSeeed Jetson One Pager
Certification
| HSCODE | 8471419000 |
| USHSCODE | 8517180050 |
| UPC | |
| EUHSCODE | 8471707000 |
| COO | CHINA |
Details
Jetson Orin Nano 8GB, 40 TOPS AI (upgradable to 67 TOPS), 7W-15W power, -10C to 60C operation, open hardware, WiFi/Bluetooth, M.2 SSD, GbE, CSI, HDMI, USB, DC powerideal for robotics, generative AI, and computer vision.
This product features a Control and UART CAN Header, an RTC 2-pin Fan Header, and more. It includes a MIPI-CSI Camera, crccq Connectors (2x), a 40-pin Expansion Header, and USB3.2 Ports (4x) with POE Gigabit Ethernet Port and LED Light. Other features include HDMI DC Power Jack and VaV Val USB Type.
Acount Area Store Manager: Unexpected incident occurred in aisle 324 at 3:30 PM. Shelf collapsed, causing boxes to fall and block the aisle.
NVIDIA Deepstream car output on NVIDIA desktop with 0.73 FPS average. Opening in blocking mode.
Jetson Orin Nano AI: Llama3, ollama, LLaVA, Llama Index, Stable Diffusion, Nanoowl, NanoDB, Whisper
Reviews
There are no reviews yet.