Seeed Studio reComputer Industrial R2245‑12 is a compact, industrial‑grade edge AI computer tailored for smart surveillance, IoT gateways, and industrial automation. Powered by Raspberry Pi CM5 and integrated with a Hailo‑8 AI accelerator , it achieves up to 26 TOPS of AI inference , enabling real‑time video analytics and intelligent monitoring at the edge.
Features
- Powerful Performance Driven by Raspberry Pi CM5: Equipped with the Arm Cortex-A76 quad-core 64-bit SoC running at 2.4GHz, it supports up to 16GB RAM, ensuring smooth multitasking and efficient data processing for AI and NVR workloads.
- Robust Fanless Heat Dissipation: The high-efficiency fanless cooling design is optimized to dissipate heat from heavy AI tasks, enabling stable operation within a wide temperature range of -20°C to 50°C.
- Industrial-Grade Reliability: Features a high-accuracy RTC, hardware watchdog, and supports optional UPS supercapacitors and TPM 2.0 security. The IP40-rated aluminum enclosure provides protection against dust and impact. Long-term availability is guaranteed until at least January 2036.
- AI-Optimized Computing Power: Features a pre-installed Hailo-8 AI accelerator delivering up to 26 TOPS, enabling low-latency, real-time AI vision processing across multiple camera channels.
- Rich Industrial Connectivity: Includes 2x RS485, 1x RS232, 4x DI/4x DO (isolated), and 1x CAN FD. Equipped with 4x Gigabit PoE PSE ports (12W max/port) for direct IP camera power supply, plus dual HDMI 2.0 and 4x USB 3.0 ports.
- Versatile Wireless Expansion: On-chip Wi-Fi and BLE; Mini-PCIe slot for LTE, LoRa®, or Zigbee; and an M.2 B-Key slot for high-speed 5G/4G connectivity.
Note:
- Options marked with * require additional purchase based on the selected bundle.
- For 4G/5G functionality, both the module and compatible antenna must be selected.
- For GPS functionality, a 4G module and GPS antenna are required.
- For LoRaWAN® functionality, a dedicated LoRaWAN® module must be selected.
Seeed Studio AI Kit Tutorial: From Zero to Hero
Master AI on Raspberry Pi with our comprehensive course. Learn to deploy computer vision tasks, including object detection, image classification, and complex visual recognition in real-world IoT applications.
Specifications
| Parameter | Description |
| Hardware Specification | |
| Product Series | R22xx-12 |
| CPU | Raspberry Pi Compute Module 5, 2.4GHz quad-core 64-bit Arm Cortex-A76 |
| AI Processor | Pre-installed 1x Hailo-8 M.2 AI Acceleration 26TOPS |
| Operating System | Raspbian, Debian |
| RAM | 8GB/16GB |
| eMMC | 32GB |
| System Specification | |
| Power Input | DC 9V~36V, 2-pin Terminal Block |
| PoE | 4 x PoE PSE |
| Power Switch | No |
| Reboot Switch | Yes |
| Interface | |
| Ethernet | 4 x 10/100/1000 Mbps(supports PoE PSE) |
| 1 x 10/100/1000 Mbps IEEE 802.3/802.3u | |
| USB | 4 x USB-A 3.0 Host |
| 1 x USB-C 2.0 (For flashing OS & Debug) | |
| RS485 | 1x RS485(Isolated) |
| RS232 | 1x RS232(Isolated) |
| CAN | 1x CAN FD(Isolated) |
| DI | 4 x Isolated DI Ports |
| Input Voltage: 5~24V DC | |
| DO | 4 x Isolated DO Ports |
| Output Voltage: <60V DC | |
| SIM Card | 1x Nano SIM Card Slot |
| M.2 Slot | 2x M.2 M-key 2280 Slot for NVMe SSD； 1x M.2 M-key 2280 Slot for AI Accelerator (Pre-installed 1x Hailo-8 M.2 AI Acceleration 26TOPS); 1x M.2 B-key 3042/3052 Slot for 5G/4G LTE |
| Mini-PCIe | 1x Mini-PCIe for LoRa module |
| LED | 4 x LED indicators |
| Buzzer | 1 |
| Reset Button | 1 |
| Audio | 1x 3.5mm Audio |
| HDMI | 2 x HDMI 2.0 |
| Wireless Communication | |
| Wi-Fi 2.4/5.0 GHz | On-chip Wi-Fi |
| BLE 5.0 | On-chip BLE |
| LoRa® | USB LoRa®*/SPI LoRa®* |
| 4G/5G Cellular | 4G LTE/5G* |
| Zigbee | USB Zigbee* |
| Standards | |
| EMC | ESD: EN61000-4-2, Level 3 |
|
| EFT: EN61000-4-4, Level 2 |
|
| Surge: EN61000-4-5, Level 2 |
| Certification | CE, FCC |
|
| TELEC |
|
| RoHS |
|
| REACH |
| Ambient Conditions | |
| Ingress Protection | IP40 |
| Operating Temperature | -20~50 °C |
| Operating Humidity | 10~95% RH |
| Storage Temperature | -40~85 °C |
| Others | |
| Supercapacitor UPS | SuperCAP UPS LTC3350 Module* |
| Hardware Watchdog | 1~255s |
| RTC | High Accuracy RTC |
| Security | Encryption Chip TPM 2.0* |
| ATECC608A | |
| Heat Dissipation | Fanless |
| Warranty | 2 years |
| Production Lifetime | Until January 2036 |
Hardware Overview
Application
This industrial AI NVR integrates advanced functions such as object detection, recognition, tracking, OCR, and abnormal event alerts. It ensures powerful edge computing performance with dependable real-time analytics, even in environments where resources are limited.
- Manufacturing & Industrial Automation
Equipped with multiple industrial ports (2x RS485, 1x RS232, 4x DI/4x DO, 1x CAN FD), it connects effortlessly to sensors, PLCs, and factory systems. Supporting a wide range of communication protocols, it enables streamlined process monitoring, automated control, and remote equipment oversight, driving smarter and more efficient production workflows.
- AI-Powered Video Analytics & Surveillance
Purpose-built as a high-performance AI-NVR, it delivers real-time multi-channel video analysis including anomaly detection, object identification, crowd density evaluation, and behavioral insights. With 4x PoE PSE Gigabit Ethernet ports (12W max each), it directly powers IP cameras, providing instant alerts and enhancing situational awareness for security operations.
- Smart Transportation
Functions as an edge AI gateway to process video and sensor inputs across multiple channels. Supports intelligent traffic flow analysis, vehicle recognition, and smart parking management. With industrial-grade durability (-20°C to 50°C operating range), it ensures consistent performance in outdoor and roadside environments, strengthening transportation monitoring and safety.
- Smart Building
Integrates with building IoT systems to manage and optimize lighting, HVAC, and access control. AI algorithms enable energy optimization, predictive maintenance, and occupant comfort adjustments. DIN-rail and wall-mount installation options allow flexible deployment in building control panels, simplifying facility management.
- Smart Retail
Provides AI-driven retail insights including product recognition, customer traffic analysis, and shopper behavior tracking. Enables personalized promotions and optimized shelf layouts. With multi-interface expansion, it integrates seamlessly with POS systems and digital signage, enhancing store efficiency and customer engagement.
Certification
| HSCODE | 8471504090 |
| USHSCODE | 8517180050 |
| UPC | |
| EUHSCODE | 8471707000 |
| COO | CHINA |
Part List
| reComputer Industrial R22xx | x 1 |
| Mounting Bracket | x 2 |
| Bracket Screw | x 4 |
| DIN-rail Clip | x 1 |
| DC Power Female Jack to Screw Terminal Adapter | x 1 |
| 19V/4.7A Power Adapter (with 1x US AC plugs + 1x EU plugs ) x1 | x 1 |
| 11*2-pin Terminal Block Connector (Male) x1 | x 1 |
| 120Ω Resistor | x 5 |
| Wi-Fi/BLE Antenna | x 1 |
| User Manual | x 1 |