Low-Power Large-Model Mainboard

Powered by the Rockchip RK3576, an octa-core 64-bit AIOT processor, AIO-3576JD4 features an advanced lithography process to deliver high
performance while maintaining low power consumption. It is equipped with an ARM Mali G52 MC3 GPU and a 6 TOPS NPU. This mainboard
supports the private deployment of ultra-large parameter models under the Transformer architecture. With support for 4K@120fps
decoding/4K@60fps encoding,it boasts a display capability with 4K resolution at a high frame rate of 120 fps. With an external watchdog,
it ensures industrial-grade reliability, making it an excellent choice for AI applications requiring local deployment.

The private deployment of large models

Octa-core 64-bit AIOT processor RK3576

RK3576, the new octa-core 64-bit AIOT processor, features a big.LITTLE architecture (4×A72 +4×A53), an advanced lithography process, and
a frequency of up to 2.2 GHz. It ensures strong support for high-performance computing and multitasking. With a Mali-G52 MC3 GPU delivering 145G FLOPS, this GPU is capable of supporting efficient heterogeneous computing to meet the demands of graphics-intensive applications.

4K@120fps high frame rate video decoding

It supports 8K@30fps/4K@120fps decoding (H.265/HEVC, VP9, AVS2, and AV1), 4K@60fps decoding (H.264/AVC), and 4K@60fps encoding
(H.265/HEVC and H.264/AVC). It provides HDMI2.1 (4K@120fps).

16MP ISP with enhanced image processing capabilities

The integrated 16-megapixel ISP supports low-light noise reduction, an RGB-IR sensor, and up to 120dB of HDR. The AI-ISP
technology enhances image quality with reduced noise. This mainboard offers support for multi-channel MIPI-CSI input.

UFS 2.0 with a data read speed of 12Gbps

With support for UFS 2.0 / HS-G3 (12Gbps), this device ensures efficient data storage and access. With the PCIe NVMe/SATA SSD expansion,
it delivers rapid data read and write speeds, enabling the device to be easily expanded to TB storage capacity.

Abundant interface options

A wide range of interface options include dual Gigabit Ethernet ports (RJ45), MIPI-CSI, HDMI 2.1, USB 3.0, USB 2.0, Type-C (Debug),
Phoenix connector (RS485, RS232, and CAN), dual-row pin headers (SPI, I2C, Line in, and Line out), M.2 (5G expansion), Mini PCIe
(4G expansion), M.2 (WiFi 6/BT 5.2 expansion), and M.2 (SATA/PCIe NVMe SSD expansion). These interfaces facilitate the
connection of various peripherals, supporting product applications across various fields.

Support various operating systems

Support Android 14, Linux OS, and Buildroot. These provide a safe and stable system environment for product research and production.

Abundant resources

We offer SDKs, tutorials, technical documentation, and development tools to streamline and improve the development process.

A wide range of applications

It is widely used in edge computing, local deployment of large models, intelligent digital signage,
cloud terminal products, industrial PCs, automotive electronics, and more.

Local deployment
of large models
Edge computing
Digital signage
Industrial PCs
Cloud terminal


Basic Specifications SOC

Rockchip RK3576


Octa-core 64-bit processor (4×A72 + 4×A53), up to 2.2GHz


G52 MC3 @ 1GHz, supporting OpenGL ES 1.1/2.0/3.2, OpenCL 2.0, Vulkan 1.1

Built-in high-performance 2D acceleration hardware


6 TOPS NPU, supporting INT4/8/16/FP16/BF16/TF32 mixed operations


The integrated 16-megapixel ISP supports low-light noise reduction, an RGB-IR sensor, and up to 120dB of HDR.

The AI-ISP enhances image quality with reduced noise.


Decoding: 4K@120fps: H.265/HEVC, VP9, AVS2, AV1; 4K@60fps: H.264/AVC

Encoding: 4K@60fps: H.265/HEVC, and H.264/AVC


LPDDR4/LPDDR4x (4GB/8GB optional)


eMMC (16GB/32GB/64GB/128GB/256GB optional), UFS 2.0 (optional)


1 * M.2 (SATA3.0/ PCIe NVMe SSD expansion, supporting 2242/2260/2280)


DC 12V (5.5mm * 2.1mm, 12V~24V wide input voltage)


Android14, Linux OS, and Buildroot


・ The private deployment of ultra-large-scale parameter models under the Transformer architecture,

such as Gemma-2B, LlaMa2-7B, ChatGLM3-6B, and Qwen1.5-1.8B

・ Traditional network architectures such as CNN, RNN, and LSTM; a variety of deep learning

frameworks including TensorFlow, PyTorch, MXNet, PaddlePaddle, ONNX and Darknet

・ Custom operator development

・ Docker container management technology


122.89mm * 85.04mm * 22.7mm




Operating temperature: -20℃~60℃

Storage humidity: 10%~90%RH (non-condensing)

Interfaces Network

Ethernet: 2 * RJ45 (1000Mbps)

WiFi: WiFi/BT module expansion available via M.2 E-KEY (2230) interface,

supporting 2.4GHz/5GHz dual-band WiFi 6 (802.11a/b/g/n/ac/ax) and BT5.2

4G: 4G LTE expansion available via Mini PCIe (shared with 5G module)

5G: 5G expansion available via M.2 (shared with 4G module)

Video Input

2 * MIPI CSI DPHY (1 * 4 lanes or 2 * 2 lanes)

1 * MIPI CSI D/C PHY (MIPI DPHY (1*4 lanes or 2 * 2 lanes) or MIPI CPHY (3 lanes) )

Video Output

1 * HDMI2.1(4K@120fps)

Audio Output

1 * 3.5mm Audio jack, supporting MIC recording, CTIA standard


External watchdog


2 * USB3.0、1 * USB2.0


1 * Type-C (USB 2.0/DEBUG), 1 * FAN (4Pin-1.25mm), 1 * SIM card

1 * dual-row pin header (2*10-20PIN-2.0mm): USB 2.0, SPI, 2*I2C, Line in, Line out, GPIO

1 * Phoenix connector (2*4Pin, 3.5mm pitch): 1 * RS485, 1 * RS232, 1 * CAN 2.0