AIBOX PRO Edge Computing Computer

AIBOX PRO Edge Computing Computer

8G+64G
$889.00
Sale price  $889.00 Regular price 
Skip to product information
AIBOX PRO Edge Computing Computer

AIBOX PRO Edge Computing Computer

$889.00
Sale price  $889.00 Regular price 
Taxes included. Shipping calculated at checkout.
Spec:8G+64G
Quantity
  • American Express
  • Apple Pay
  • Google Pay
  • Mastercard
  • PayPal
  • Shop Pay
  • Union Pay
  • Visa
Brand Firefly
Availability 9 in stock

AIBOX PRO Edge Computing Box

AIBOX PRO is a compact host designed for edge large model deployment. It supports LLM, VLM, AI Agent, and multimodal large model implementation, while also remaining compatible with traditional algorithms such as CNN and RNN. The device adopts a heterogeneous modular architecture featuring a main controller plus an independent AI coprocessor, both designed as modules. This allows flexible selection of modules with different computing power and memory specifications based on application requirements. The device integrates CAN-FD and RS485 industrial interfaces, making it suitable for both general AI and industrial intelligence deployment needs.

Dual-Core Architecture

Heterogeneous primary + secondary modular architecture connected via PCIe. Primary module: task scheduling & data preprocessing. Secondary module: core AI inference, model parallelism, multi-task coordination. Ideal for robot navigation and multimodal perception — efficient resource allocation, low latency, stable operation for complex tasks.

Primary Module 8K Codec support

Primary module integrates a high-performance VPU — 8K hardware encode/decode, plus hardware-accelerated scaling and cropping. Handles multi-channel 4K video streams in parallel for efficient edge video preprocessing. Flexible hardware, storage, and OS options for AI inference runtime. Rich I/O interfaces for HMI, data acquisition, and real-time control — expanding application possibilities.

High-Compute Modules for Needs

The secondary module supports high-compute, high-bandwidth modules such as RK1828 and HOUMO LQ50, providing powerful computing foundation for scenarios including LLMs, VLMs, MoE models and AI Agents. Its compute-in-memory architecture significantly increases DRAM bandwidth, improving LLM inference TPS and time-to-first-token performance. Future upgrades will support RK1860.

Millisecond-Level Low Latency

With local AI processing, network dependency is minimized — no need to upload data to the cloud, and operations are unaffected by network bandwidth or speed fluctuations, enabling millisecond-level response and data privacy protection. All business data is stored and processed locally, meeting data security regulations and industry privacy compliance requirements.

Onboard PCle Interface for Storage

Onboard M.2 PCIe high-speed interface supports NVMe SSD expansion. Optional high-speed SSDs can be added to easily expand total system storage to terabyte level, meeting local storage requirements for large model packages and massive business data.

Full Aluminum Alloy Shell

Equipped with an industrial-grade full aluminum alloy enclosure and an active cooling fan for efficient heat dissipation. Supports 24/7 uninterrupted stable operation, maintaining excellent computing performance and system stability even under high-temperature conditions — fully adaptable to the needs of various industrial scenarios.

Rich Expansion Interfaces

Offers a variety of interfaces including dual Gigabit Ethernet, USB 3.0, HDMI, NVMe, 4G/5G, RS485, CAN-FD, and opto-isolated DI/DO, flexibly adapting to various industrial peripherals and industry application expansions.

Provides Development Materials

Provides accompanying source code, tutorials, technical documentation, and development tools, making development easier and more convenient.

M.2 Interface for Diverse AI Module Expansion

Adopts heterogeneous modular architecture featuring a main controller + independent AI coprocessor. The main controller is compatible with flagship SoCs including Rockchip RK3588, SpacemiT K3, and Qualcomm QCS8550. Dual M.2 slots can accommodate two accelerator cards simultaneously, supporting modules such as Rockchip RK1828, HOUMO LQ50, and DeepX DX-M1, enabling capacity expansion without hardware modification. Equipped with network ports, video storage, as well as industrial interfaces including RS485, CAN-FD and opto-isolated DI/DO, it is well-suited for industrial deployments across various sectors.

Application Scenarios

Widely applicable in industries such as robotics, AI edge computing, industrial automation, intelligent security, autonomous driving, smart cities, traffic management, home storage, and retail automation.

Robots
Edge Computing
Industrial Automation
Industrial Control Host
Smart City
Traffic Management

Interfaces

Specifications

AIBOX PRO
Basic Specifications Host module

Optional modules as follows:
RK3588 module (Core-3588JD4): LPDDR4/LPDDR4x (4/8/16/32GB) + eMMC (32/64/128/256GB), 6TOPS(INT8)
RK3576 module (Core-3576JD4): LPDDR4/LPDDR4x (2/4/8/16GB) + eMMC (16/32/64/128/256GB), 6TOPS(INT8)

Video encoding
RK3588: 8K@30fps H.264
RK3576: 4K@60fps H.264
Video decoding

RK3588: 8K@60fps VP9/AVS2, 8K@30fps H.264 AVC/MVC, 4K@60fps AV1, 1080P@60fps MPEG-2/-1/VC-1/VP8
RK3576: 8K@30fps VP9/AVS2/AV1, 4K@120fps VP9/AVS2/AV1, 4K@60fps H.264/AVC

Accelerator expansion (Slave module)

2 × M.2 M-KEY (Supports 2260/2280 accelerator cards & other M.2 modules, inside the device)
* Users can install any brand M.2 M-KEY accelerators; standard equipped with RK1828/HOUMO LQ50 modules
○ RK1828: Onboard 5GB DRAM, 20TOPS(INT8), runs LLMs up to 7B smoothly
○ LQ50: 48GB LPDDR5, 160TOPS(INT8), runs LLMs up to 35B smoothly

Storage expansion

1 × M.2 M-KEY (Expandable with PCIe2.0 x1 NVMe 2280 SSD; unavailable when core-3576JD4 serves as the host module)

1 × SATA3.0 (Expandable with SATA3.0 SSD; unavailable when core-3576JD4 serves as the host module)

1 × TF Card

OS

Linux

Device power

DC 12V/7A (5.5×2.1mm, support 9V~36V wide voltage input)

Accelerator power

2 × 12V (2pcs 4Pin-1.25mm wafer sockets)

Power consumption

RK3588 + 1 × RK1828: Normal: 9.6W(12V/800mA), Max: 28.8W(12V/2400mA), Min(Sleep): 0.72W(12V/60mA)

RK3588 + 2 × RK1828: Normal: 15.6W(12V/1300mA), Max: 42W(12V/3500mA), Min(Sleep): 0.72W(12V/60mA)

RK3588 + 1 × LQ50: Normal: 8.4W(12V/700mA), Max: 45.6W(12V/3800mA), Min(Sleep): 0.84W(12V/70mA)

RK3588 + 1 × LQ50 DUO: Normal: 12W(12V/1000mA), Max: 63.6W(12V/5300mA), Min(Sleep): 0.84W(12V/70mA)

RK3576 + 1 × RK1828: Normal: 8.4W(12V/700mA), Max: 28.8W(12V/2400mA), Min(Sleep): 1.2W (12V/100mA)

Dimension

160mm × 111mm × 61mm

Environment

Operating Temperature: -20℃~60℃, Storage Temperature: -20℃~70℃, Storage Humidity: 10%~90%RH (non-condensing)

Interface Specifications Internet

2 × Gigabit Ethernet (1000Mbps/RJ45)
Wi-Fi & Bluetooth: Support dual-band 2.4GHz/5GHz Wi-Fi and Bluetooth 5.2 (Expandable via M.2 E-KEY inside the device; unavailable with core-3576JD4 host module)
4G: Expandable via Mini PCIe inside the device
5G: Expandable via M.2 B-KEY inside the device

Audio output

1 × Line In (3.5mm audio jack), 1 × Line Out (3.5mm audio jack)

USB

2 × USB3.0 (Max: 1A), 1 × Type-C (USB2.0 OTG), 1 × Type-C (Debug)

Antenna

2 × Wi-Fi Antenna, 1 × 4G Antenna

Button

1 × Power key, 1 × Upgrade key

Other interfaces

1 × SIM Card

Customization

Firefly team provides hardware, software, turnkey and OEM/ODM customization services.