Resources:
Product Overview
Radxa Dragon Q8B Snapdragon 8cx Gen 3 SBC with 29+ TOPS AI Engine, Dual 2.5GbE and Triple 4K120 Display
The Radxa Dragon Q8B is a high-performance single-board computer and compact motherboard designed for edge AI, robotics, intelligent terminals, multimedia processing, development workstations, and embedded computing. Powered by the 5 nm Qualcomm Snapdragon 8cx Gen 3 (SC8280XP), it combines an 8-core 64-bit Kryo CPU with a Qualcomm Adreno 690 GPU and a Qualcomm Hexagon processor. The Qualcomm AI Engine provides up to 29+ TOPS of combined AI acceleration, making the board suitable for computer vision, object detection, image enhancement, local inference, intelligent automation, and other edge-computing workloads supported by the Qualcomm software stack.
Dragon Q8B supports up to 32 GB LPDDR4X memory and provides flexible high-speed storage through 2 M.2 M-Key 2280 slots, a UFS module connector, and microSD. Dual 2.5GbE ports, USB 3.2 Gen 2, HDMI 2.1, DisplayPort 1.4 over USB-C, a 40-pin GPIO header, and M.2 E-Key wireless expansion make it suitable for both desktop-class and embedded deployments. Support for Radxa OS, Ubuntu, Debian, Windows 11 ARM64, and other Arm Linux distributions provides developers with a flexible software environment for edge intelligence, application development, and general-purpose computing.
Key Features
- Snapdragon 8cx Gen 3 Platform: Powered by the Qualcomm SC8280XP 5 nm compute platform for high-performance Arm computing.
- 8-Core Kryo CPU: Integrates 4 Kryo Prime cores at up to 3.0 GHz and 4 Kryo Gold cores at up to 2.4 GHz.
- 64-Bit Architecture: Provides modern Arm-based computing capability for Linux, Windows on Arm, and embedded applications.
- Qualcomm Adreno 690 GPU: Delivers accelerated graphics processing with DirectX and DirectML support.
- 29+ TOPS AI Engine: Qualcomm AI Engine provides up to 29+ TOPS of combined AI compute performance for supported edge-inference workloads.
- Qualcomm Hexagon Processor: Includes Hexagon acceleration resources for AI, DSP, and sensing workloads.
- Up to 32 GB LPDDR4X: Supports high-capacity memory for AI applications, multitasking, development, and multimedia processing.
- 4266 MT/s Memory: LPDDR4X memory operates at data rates up to 4266 MT/s across a 128-bit memory interface.
- 4K120 Video Decode: Integrated Adreno VPU supports video decoding at resolutions up to 4K120.
- 4K60 Video Encode: Supports hardware video encoding up to 4K60 for compatible codecs.
- Triple 4K120 Display Capability: Provides HDMI 2.1 plus 2 USB-C DisplayPort 1.4 Alt Mode outputs for multi-display systems.
- Dual 2.5GbE: 2 RJ45 Ethernet ports provide high-speed wired networking for edge servers, robotics, gateways, and data-intensive applications.
- Dual NVMe Expansion: Includes 2 M.2 M-Key 2280 slots with PCIe 3.0 x4 and PCIe 3.0 x2 connectivity.
- UFS 3.1 Support: Dedicated UFS module connector provides high-performance embedded storage expansion.
- microSD Support: Onboard microSD slot provides convenient removable boot and storage capability.
- Wireless Expansion: M.2 E-Key 2230 slot supports compatible Wi-Fi and Bluetooth modules.
- USB 3.2 Gen 2: Includes 2 USB-C and 2 USB-A high-speed USB 3.2 Gen 2 ports.
- Additional USB 2.0: 2 USB 2.0 Type-A ports support keyboards, mice, serial adapters, and other peripherals.
- 40-Pin GPIO Header: Exposes UART, I2C, SPI, and GPIO interfaces for sensors, control electronics, and embedded hardware.
- PCIe FPC Expansion: Provides an additional PCIe 3.0 x1 interface through an FPC connector.
- Audio Connectivity: Includes a 3.5 mm headphone jack and a dedicated microphone connector.
- Active Cooling Support: Dedicated fan connector simplifies thermal management under sustained processing loads.
- RTC Connector: Supports an external RTC battery for maintaining system time when main power is removed.
- Multiple Operating Systems: Supports Radxa OS, Ubuntu, Debian, Armbian, Arch Linux, NixOS, Windows, and other compatible Arm operating systems.
- Compact 100 x 75 mm Form Factor: Combines extensive desktop-class connectivity with a compact embedded-board footprint.
Applications
- Edge AI inference
- Computer vision
- Object detection
- Image classification
- Image enhancement
- Robotics
- Autonomous systems
- Intelligent terminals
- Industrial edge computing
- AI gateways
- Smart kiosks
- Digital signage
- Human-machine interfaces
- Multimedia systems
- Network appliances
- Development workstations
- Embedded Linux development
- Windows on Arm development
- Machine-learning prototyping
- Local AI applications
Interfaces
