The VCK5000 development card is built on the Xilinx 7 nm Versal ACAP architecture and is designed for designs requiring high throughput AI inference and signal processing compute performance. For this release, DPU core with batch=4, batch=6, and batch=8 are implemented using AI Engines. In theory, each processing engine can provide 10 TOPS INT8 peak performance for deep learning inference, and about 7.5 TOPS INT8 peak performance after the efficiency loss of the AI Engine compiler.