Cubie A7A

The New Standard for Affordable Edge AI Innovation

Buy Now

Radxa Cubie A7A

2× Cortex-A76 +
6× Cortex-A55
Imagination
BXM-4-64 GPU
WiFi 6
BT 5.4
Gigabit Ethernet
LPDDR5 RAM
up to 16GB
Support eMMC / UFS
Module
PCIe 3.0

Exceptional Performance, Flagship Experience

High-performance hybrid
octa-core SoC

CPU: 2× Cortex-A76 + 6× Cortex-A55, The
maximum main frequency is 2.0 GHz.

GPU: Imagination BXM-4-64
MC1, Efficient Graphics
and Rendering

NPU: 3 TOPS@INT8, Integrated
Vivante VIP9000 NPU, native INT8 /
INT16 / FP16 / BF16, multi-precision
mixed computing

MCU: RISC-V E902 (200MHz) Coprocessor, Real-time tasks and low power management

Unbeatable Value, Developer’s Favorite

Maximum power at minimum cost — a true game-changer for makers, developers, and system integrators.

PRICE
Half the price of similar products, with an extremely high TOPS / $ performance ratio
RAM
Starting from $25 with 2GB ram, up to 16GB LPDDR5 to meet large model inference needs
STORAGE
eMMC / UFS / NVMe multi-storage solutions with capacity up to 1TB

Rich Interfaces for Limitless Expansion

Packed with PCIe, HDMI, MIPI, USB, and Gigabit Ethernet with PoE support — ready for any project.

Advanced Wireless Connectivity

Seamless connectivity with Wi-Fi 6, Bluetooth 5.4, and Gigabit Ethernet with optional PoE.

Wi-Fi 6（802.11ax）+ Bluetooth
5.4 + BLE

Gigabit wired network, supporting
IEEE 802.3af/at PoE

Flexible Power Options with PoE

Flexible power options including USB-C, GPIO, and PoE, making it easy to deploy in industrial scenarios.

Superior Multimedia Experience

Delivering crisp 4K video output and supporting a wide range of camera and display accessories.

Support:

- Encoding: 4K@30fps H.264/H.265
- Decoding: 4K@60fps H.265 / VP9 / AVS2, 4K@30fps H.264
- Supports dual-display (HDMI + MIPI DSI)
- Supports CEC / HDCP 1.4

Commercial-Grade Reliability and Longevity

Credit Card Size (85×56mm), Easy to Integrate
High-quality Components, Hardware Encryption Support
Guaranteed Supply Availability through 2034

Complete AI Development Ecosystem

From model import to cross-platform deployment — in minutes

Compatible with all major AI frameworks

Cross-platform deployment

Linux/Android universal NBG binary

10 minutes

Run Yolov8, Mobile-SAM, Whisper instantly

Zero-Copy API
Custom Operator
Multi-thread

RISC-V E902 Always-On Co-Processor

Low Power · Instant Response · Secure by Design

Open Architecture
GCC/LLVM, VS Code toolchain support
Ultra low Power Consumption
Runs tasks independently while main core sleeps
Security
Secure Boot & Root-of-Trust verification

Real-time
For motor control and sensor data collection
Cost Optimization
Lower BOM by $1-2
Direct Peripheral Connection
Supports Multiple Interfaces

Multiple OS Support for Flexible Development

Open software environment supporting Debian Linux and Android, ideal for both rapid prototyping and deployment.

Supporting a Variety of Accessories
to Create a Complete Ecosystem

Typical Applications

smart home
smart city
Industrial Internet of Things
Agricultural Internet of Things
smart retail
intelligent medical care
STEM Education
Video Conference
unmanned aerial vehicle
Robotics
AR / VR
intelligent transportation

Comprehensive Technical Documentation and
Active Community

Comprehensive Technical Documentation and Active Community !
Radxa provides comprehensive technical documentation, empowering users to fully utilize their products. At the same time, they actively maintain an engaged community, fostering communication and collaboration among users.

The terms HDMI, HDMI High-Definition Multimedia Interface, HDMI Trade dress and the HDMI Logos are trademarks or registered trademarks of HDMI Licensing Administrator, Inc.