Orion O6N
Product Overview
The Radxa Orion O6N is a high-performance Nano-ITX (120mm x 120mm) motherboard based on the CIX P1 SoC, featuring a powerful 12-core CPU architecture, Arm Immortals™ G720 MC10 GPU, and a 30 TOPS NPU.
This product supports up to 64GB LPDDR5 memory and multiple storage expansion options, providing rich video input/output interfaces, high-speed network connectivity, and comprehensive expansion capabilities.
The O6N is specifically designed for high-performance applications such as AI inference, graphics processing, video codec, and multi-display, making it an ideal choice for edge computing, smart devices, and multimedia applications.
Product Appearance
| No. | View | No. | View | No. | View |
|---|---|---|---|---|---|
| ① | Radxa Orion O6N Front View | ② | Radxa Orion O6N Side View | ③ | Radxa Orion O6N Rear View |
Package Contents
- Radxa Orion O6N Motherboard
- Radxa 8420B Heatsink
- Radxa DC12 60W Power Adapter
Note: The actual package contents are subject to the purchase platform's specifications.
Product Specifications
| Category | Specifications |
|---|---|
| System on Chip (SoC) | CIX P1 |
| Central Processing Unit (CPU) | 4x Big Cores Cortex®-A720, up to 2.6 GHz 4x Middle Cores Cortex®-A720, up to 2.4 GHz 4x Small Cores Cortex®-A520, up to 1.8 GHz System Cache: 12MB L3 cache shared across all cores |
| Graphics Processing Unit (GPU) | Arm® Immortals™ G720 MC10 - Hardware-accelerated ray tracing - OpenGL ES 3.2 support - Vulkan 1.3 support - OpenCL 2.2 support - DirectX 12 support |
| Neural Processing Unit (NPU) | 30 TOPS computing power - Supports INT4 / INT8 / INT16 / FP16 / BF16 / TF32 precisions |
| Video Codec | Decoding: - Supports AV1 / H.265 / H.264 / VP9 / VP8 / H.263 / MPEG-4 / MPEG-2 formats - Up to 8K@60fps decoding Encoding: - Supports H.265 / H.264 / VP9 / VP8 formats - Up to 8K@30fps encoding |
| Memory | LPDDR5 - 128-bit data bus width - Transfer speed: 5000 MT/s or 6000 MT/s options - Capacity options: 8GB / 16GB / 24GB / 32GB / 48GB / 64GB |
| Storage | 1x UFS module interface for Radxa UFS module expansion 2x M.2 M Key 2280 slots (PCIe Gen4 x4) for M.2 NVMe SSD expansion |
| Video Output | Supports up to 3 simultaneous displays - 1x USB Type-C with DP Alt Mode, up to 4Kp60 - 1x HDMI port, up to 4Kp60 - 1x DP port with Multi-Stream Transport, up to 4Kp120 |
| Video Input | 2x 4-lane MIPI CSI interfaces |
| USB Ports | 1x USB 3.2 HOST Type-C 2x USB 3.2 HOST Type-A 3x USB 2.0 HOST Type-A |
| Networking | 2x 2.5G Ethernet 1x M.2 E Key 2230 for Wi-Fi and Bluetooth 1x M.2 B Key 3042 for mobile network (4G) 1x Nano SIM card slot for mobile network (4G) |
| Power Supply | 12V DC power input or 12V pin power input |
| Cooling | 4-Pin fan header with PWM speed control and tachometer 4x heatsink mounting points - Mounting holes: M3.5 - Dimensions: 75mm x 75mm |
| Other Interfaces | 1x Power button 1x 40-Pin GPIO header 1x RTC battery holder (CR1220 compatible) 1x 4-Pin power interface: - 12V supports input/output - 5V supports output only |
| Recommended Power Supply | ≥ 65W |
| Dimensions | Nano-ITX: 120mm x 120mm |
Interface Specifications
| No. | Description | No. | Description | No. | Description |
|---|---|---|---|---|---|
| 1 | USB Type-C Port (Supports DP Video Output) | 9 | 12V DC Power Input (5525) | 17 | M.2 B Key 3042 Slot |
| 2 | 4-Pin Fan Header (PWM Speed Control & Tachometer) | 10 | Standard DP Port | 18 | Nano SIM Card Slot |
| 3 | 40-Pin GPIO Header | 11 | Standard HDMI Port | 19 | CIX P1 SoC |
| 4 | MIPI CSI Interface (4-lane) | 12 | 2x USB 3.0 Type-A | 20 | SPI Nor Flash (BIOS) |
| 5 | LPDDR5 | 13 | USB 2.0 Type-A | 21 | UFS Module Interface |
| 6 | M.2 E Key 2230 Slot | 14 | Power Button | 22 | RTC Battery Connector |
| 7 | MIPI CSI Interface (4-lane) | 15 | 5V / 12V Power Port | 23 | 2x 2.5G Ethernet |
| 8 | M.2 M Key 2280 Slot | 16 | M.2 M Key 2280 Slot | 24 | 2x USB 2.0 Type-A |
Application Scenarios
AI Inference and Machine Learning
- Intelligent Visual Analysis: Leverage 30 TOPS NPU computing power for real-time image recognition, object detection, and behavior analysis
- Natural Language Processing: Support localized AI model deployment, providing efficient speech recognition and natural language understanding capabilities
- Edge AI Computing: Run complex machine learning models on edge devices, reducing cloud dependency
Education and Research
- AI Teaching Platform: Provide students with a practical environment for local AI model training and inference
- Scientific Computing: Support deep learning research, computer vision algorithm development, and data analysis
- Prototype Validation: Quickly validate AI algorithms and models to accelerate research outcomes
Robotics Development
- Autonomous Navigation: Combine visual processing and AI inference to achieve environmental perception and path planning for robots
- Robotic Arm Control: Utilize real-time computing capabilities to support precise robotic motion control and visual servoing
- Human-Robot Interaction: Enable natural human-robot interaction through speech recognition and visual analysis
Multimedia Processing
- 8K Video Processing: Support 8K@60fps decoding and 8K@30fps encoding, suitable for professional video editing and streaming applications
- Multi-Display System: Support for three simultaneous displays up to 4Kp120, ideal for digital signage, monitoring centers, and multimedia displays
- Graphics Rendering: Arm Immortals™ G720 GPU with hardware-accelerated ray tracing, suitable for 3D rendering and game development
Industrial and Commercial Applications
- Industrial Automation: Connect industrial cameras via MIPI CSI interfaces for quality inspection and production monitoring
- Smart Retail: Integrated AI vision analysis supports customer flow statistics, product recognition, and intelligent recommendations
- Digital Signage: Multi-display capability is ideal for advertising, information display, and interactive kiosks
Networking and Communication
- Gateway Devices: Dual 2.5G Ethernet ports and 4G network support make it suitable as an industrial gateway or router
- Remote Monitoring: Combine video processing and network capabilities to build remote monitoring and video conferencing systems
- IoT Edge Node: Rich interfaces and powerful processing capabilities make it ideal as an IoT data aggregation and processing node
Software Support
- Debian / Ubuntu Linux distributions
- Full UEFI support (via EDKII implementation)
- Hardware adaptation package (BSP) and software development kit (SDK) provided