Xuanyuxin (abbreviated as Xuanyuxin) is a fully self-developed LLM inference card designed for efficient reasoning of large-scale language models.

The product integrates self-developed LPU, and achieves high-performance, low-power AI reasoning through software and hardware collaborative optimization. The original intention of research and development is to accelerate the low-cost implementation of AI applications in different industries through ultra-high cost performance.

The high-performance edge computing board based on the Xilinx Zynq UltraScale+ MPSoC ZU9EG chip integrates multi-core processors, powerful reconfigurable logic resources, rich storage and high-speed IO interfaces, and multimedia processing capabilities. It is designed for edge computing scenarios with strict requirements on real-time and computing power, such as 5G communications, video processing, intelligent driving, and industrial vision.

The AI ​​computing hardware platform with RISC-V 64-bit multi-core CPU and deeply customized DPU as the core has the characteristics of low power consumption, high computing power, and strong ecology. Combined with rich I/O interfaces and industrial-grade design, it launches professional models for scenarios such as audio processing, visual recognition, and live edge computing, providing a one-stop AI development and deployment solution.