APLUX

Compact edge AI development board

Rhino Pi A1 Mini

Bring edge AI into compact devices

Built on Qualcomm Dragonwing QCS6490 and the Rhino Core Q64 module, with 8 GB LPDDR5 and 64 GB eMMC. Connect dual cameras, displays, networks and control peripherals on an 85 × 56 mm board for visual recognition, touch interaction and robot perception development.

Rhino Pi A1 Mini front view with Q64 module

12 TOPS

Maximum AI performance (INT8, dense)

8 GB + 64 GB

LPDDR5 / eMMC 5.1

85 × 56 mm

Reference PCB outline

Dual CSI

MIPI camera interfaces

Dimensions refer to the PCB outline. Allow space for connectors, the compute module and heatsink. AI performance is the platform's nominal rating.

A compact board for vision and interaction

Local compute, image capture, displays and peripheral connections on one board for compact device development.

Compact 85 × 56 mm board

The Q64 module and compact carrier support mechanical and system integration in space-constrained terminals.

12 TOPS of local AI compute

Qualcomm Dragonwing QCS6490 combines an octa-core Kryo 670 CPU, Adreno 643 GPU and up to 12 TOPS (INT8, dense) AI performance for object detection, image classification and other edge applications.

Dual-camera capture and display interaction

Dual 4-lane MIPI CSI, Micro HDMI, MIPI DSI and I²C touch interfaces connect image inputs with terminal displays.

Connect networks and control peripherals

Gigabit Ethernet, Wi-Fi 6, Bluetooth 5.4, four USB-A ports and a 40-pin header support networking, peripheral access and bus integration.

From interface capabilities to device integration

Locate connectors on the hardware and plan device connections around your peripherals.

Rhino Pi A1 Mini front

Select a number or interface name for connection details. Refer to the specification for pin assignments, voltage levels and multiplexing.

12 V power input

J6 USB-C and J12 power header are alternative 12 V inputs and must not be connected simultaneously. Recommended supply: 12 V / 2 A. USB-C is not a USB data port.

Explore integration paths for real-world tasks

Choose device workloads and integration paths based on platform capabilities and this model's interfaces.

Qualcomm Dragonwing QCS6490

Bring vision and interaction to industry terminals

Visual recognitionTouch interactionPeripheral connectivity
Up to 12 TOPS(INT8, dense)Platform-rated peak; validate performance for your workload
Illustrative industrial handheld terminal

Industrial handheld

Combine touch interaction, image capture and local recognition in portable terminals for inspection, warehousing and field operations.

Platform fit
Combine local vision processing with touch interaction to bring recognition results directly into field workflows.
Board connections
The compact 85 × 56 mm carrier supports prototyping; scanners, display, battery and enclosure require system-level design.
Field captureLocal recognitionOperational feedback
View interfaces and integration
Illustrative smart camera

Smart Cameras

Integrate capture, recognition and result output on the device for detection, counting and status recognition tasks.

Platform fit
Combine the ISP, image processing and AI inference into a capture-to-decision vision pipeline.
Board connections
Connect cameras through dual MIPI CSI, then use GPIO and networking for recognition and status feedback.
Image captureVision inferenceResult output
View interfaces and integration
Illustrative industry tablet

Industry tablet

Connect displays, touch input, cameras and business systems to support industry user interfaces and local intelligent interaction.

Platform fit
Combine cameras, voice and business applications in a display-equipped device to create an actionable industry interface.
Board connections
Connect displays over HDMI / MIPI DSI and touch peripherals over USB. Connector positions are shown in this page's hardware photos.
Touch inputContent understandingBusiness interaction
View interfaces and integration
Illustrative retail terminal

Retail terminal

Combine scanning, displays and networking for checkout, self-service and in-store information terminals.

Platform fit
Run scanning, image recognition and interface feedback on the terminal and connect them to store workflows.
Board connections
Connect scanners over USB, then combine display and networking to prototype self-service interactive terminals.
Scanning and inputBusiness processingService interaction
View interfaces and integration

Device images illustrate application directions; peripherals, enclosure design and software require project-specific integration.

Rhino Pi A1 Mini specifications

Configurations below follow the compact edge AI development board documentation. Module capabilities and physically exposed board interfaces are distinguished.

Compute and storage

Rhino Pi A1 Mini · Compute and storage
Product modelRhino Pi A1 Mini
Compute moduleRhino Core Q64 module
Compute platformQualcomm Dragonwing QCS6490
Process6 nm
CPUOcta-core Qualcomm Kryo 670, up to 2.7 GHz
NPUQualcomm Hexagon 770
AI performance (INT8, dense)Up to 12 TOPS(INT8, dense)
GPUQualcomm Adreno 643, up to 812 MHz
Memory8 GB LPDDR5
System storage64 GB eMMC 5.1
Storage expansion1 × Micro SD / TF slot; SD v3.0, backward compatible with 2.0

AI figures are platform-rated peaks. Dense, sparse and FP16 figures are listed separately; actual performance depends on the model, precision, software and cooling. Input power ratings do not represent total board consumption. Accessories and system images depend on the delivered configuration.

85 × 56 mm is the reference PCB outline and excludes complete assembly height. Check protruding connectors, the compute module and heatsink; refer to the companion DXF for detailed dimensions. Input power ratings do not represent total board consumption. Use a validated supply and cable delivering 12 V and at least 2 A, and check total peripheral load. Validate operating temperature against system thermal design and workload. The base board includes the Q64 module and compact carrier. The order configuration determines whether a heatsink, supply, antennas, debug cable, PoE HAT, display, cameras and PCIe peripherals are included. Video codec figures describe chipset capabilities; actual performance depends on the system image and application pipeline, not HDMI output specifications or concurrent dual-camera frame rates.

Start your next compact AI device with A1 Mini

Share your camera, display, peripheral, mechanical and system requirements so we can confirm the board, image and accessory configuration.