Description Waveshare Luckfox Core3576 Module is a high-performance AIoT core module built around the Rockchip RK3576 octa-core processor. It combines 4 x Cortex-A72, 4 x Cortex-A53, a built-in 6 TOPS NPU, and a rich multimedia subsystem in a compact 260-pin SO-DIMM form factor, making it suitable for AI edge computing, industrial control, intelligent display systems, machine vision, robotics, and advanced embedded Linux products. This configuration provides 8GB LPDDR4X RAM and 64GB eMMC Flash, along with support for dual Gigabit Ethernet, Wi-Fi 6, Bluetooth 5.2, MIPI CSI, MIPI DSI, HDMI 2.1, eDP 1.3, PCIe, SATA, and other high-speed interfaces. For projects that need a compact but powerful Rockchip core module, the Luckfox Core3576 module is positioned as a compute engine rather than a simple entry-level board. Features RK3576 octa-core processor with 4 x Cortex-A72 and 4 x Cortex-A53 CPU architecture. 6 TOPS NPU for AI inference and edge model acceleration. 8GB LPDDR4X RAM + 64GB eMMC Flash for demanding embedded workloads. Mali G52 MC3 GPU for graphics acceleration and multimedia output. 4K video decode and encode support for smart display, vision, and media applications. Dual Gigabit Ethernet plus onboard Wi-Fi 6 and Bluetooth 5.2. Dual MIPI CSI camera interfaces for AI vision and image capture systems. HDMI 2.1, eDP 1.3, and MIPI DSI for multi-display integration. 260-pin SO-DIMM connector with rich expansion interfaces including PCIe, SATA, USB, I2C, SPI, UART, and GPIO. System support for Buildroot, Debian, and Android 14. Specifications Item Details Processor RK3576 octa-core (4 x Cortex-A72 + 4 x Cortex-A53) RAM 8GB LPDDR4X Flash 64GB eMMC NPU 6 TOPS GPU Mali G52 MC3 @ 0.9GHz ISP Built-in 16MP ISP with HDR, 3A, CAC, 3DNR and related image processing support Video Decode 4K@120fps (H.265/HEVC, VP9, AVS2, AV1), 4K@60fps (H.264/AVC) Video Encode 4K@60fps (H.265/HEVC, H.264/AVC) Connectivity Dual Gigabit Ethernet, Wi-Fi 6, Bluetooth 5.2 Video Output HDMI 2.1 (4K@120fps), eDP 1.3 (4K@60fps), MIPI DSI (2K@60fps) Camera Input Dual MIPI CSI USB USB3.2 Gen1 OTG, USB3.2 Gen1, USB2.0 Expansion PCIe 2.1, SATA3.1, SDIO3.0, CAN FD, I2C, I3C, SPI, UART, GPIO Connector 260-pin SO-DIMM Power 5V DC Dimensions 69.6mm x 45mm System Buildroot, Debian, Android 14 Applications AI edge computing for local model inference and intelligent device deployment. Industrial control systems requiring a powerful Linux compute core with stable IO expansion. Smart displays and multimedia terminals that benefit from 4K video pipelines and multi-display support. Robotics and machine vision using dual camera interfaces and NPU acceleration. Automotive and mobile intelligent systems that need a compact high-performance core module. Custom carrier-board based development using the 260-pin SO-DIMM connector. Product List Core3576 module x1 Other Content 6 TOPS AI acceleration is one of the biggest reasons to choose the Luckfox Core3576 module over simpler Linux boards. It gives this RK3576 module a much stronger position in AI vision, recognition, and edge inference projects where CPU-only solutions are not enough. Three-screen display capability is another key strength. With HDMI 2.1, eDP 1.3, and MIPI DSI, this module can support more advanced display products and industrial HMI designs than simpler embedded compute modules. Dual Gigabit Ethernet and rich high-speed interfaces make this module much easier to integrate into gateways, AI boxes, robotics, and industrial controllers built around a custom carrier board. Resources Luckfox Documentation Center Luckfox GitHub Repository Core3576 Datasheet RK3576 Datasheet RK3576 Reference Manual FAQ Is this a complete development board or a core module?This SKU is the Luckfox Core3576 core module in a 260-pin SO-DIMM form factor, intended to work as the compute core in larger system designs. What memory and flash configuration does this version provide?This version provides 8GB LPDDR4X RAM and 64GB eMMC Flash. What makes this module suitable for AI edge computing?It combines the RK3576 octa-core processor with a built-in 6 TOPS NPU, which is much better suited to AI inference than basic Linux modules. Can it support multiple display outputs?Yes. The official specification includes HDMI 2.1, eDP 1.3, and MIPI DSI, making it suitable for advanced multi-display products.