On October 28, Qualcomm announced the launch of its next-generation AI inference-optimized solutions for data centers, including accelerator cards and rack systems based on the Qualcomm AI200 and AI250 chips. Leveraging its accumulation in NPU technology, this series of solutions focuses on rack-level performance and memory capacity optimization. It can support generative AI inference with high efficiency, helping various industries advance scalable and high-efficiency AI deployment.