CQ38W-3576 AI Camera

Equipped with an octa-core 64-bit RK3576 AIOT processor, big.LITTLE architecture (4×A72 + 4×A53), up to 2.2 GHz, delivering powerful performance for computing and multitasking. Features a Mali-G52 MC3 GPU with 145 GFLOPS for efficient heterogeneous computing and graphics-intensive applications.

Supports private deployment of ultra-large-scale parameter models under the Transformer architecture, such as Gemma, ChatGLM, Qwen, Phi and other large language models. Also supports traditional network architectures including CNN, RNN, LSTM, RKNN model import/export, and various deep learning frameworks such as TensorFlow, TensorFlow Lite, PyTorch, Caffe, ONNX, and Darknet.

Built-in 16MP ISP with support for RGB-IR sensors, featuring 120dB HDR, 3A, CAC, 3DNR, sharpening, dehazing, and other image enhancement and correction algorithms. AI-ISP improves low-noise image quality.

Supports 8K@30fps / 4K@120fps (VP9, AVS2, AV1) and 4K@60fps video decoding (H.264 / AVC), as well as 4K@60fps video encoding (H.264 / AVC).

Optional 5MP / 3MP lenses with high sensitivity, high signal-to-noise ratio, and low illumination. High-definition image quality, accurately capturing highlight and shadow details in dynamic scenes, with support for infrared night vision.

Equipped with an adaptive 100M Ethernet port that intelligently negotiates network speed, delivers stable and smooth transmission, and features excellent anti-interference performance and link reliability — easily adapting to complex network environments such as security surveillance.

Supports RTLinux kernel with excellent real-time performance, widely used in industrial application scenarios. Supports operating systems such as Android 14, Linux OS, and Buildroot, providing a secure and stable system environment for product development and production.

Adopts an IP67 waterproof enclosure with a built-in heatsink, weighing under 350g, and supports various mounting methods such as wall mounting and ceiling mounting.