Shares of Qualcomm Technologies, Inc. soared 11 percent on Monday after the company unveiled two artificial intelligence accelerator chips for data centers—a bold leap beyond the chipmaker’s traditional stronghold in mobile and consumer semiconductors.
The Qualcomm rally placed it among the top gainers in the S&P 500, underscoring investor enthusiasm its expansion into the red-hot AI infrastructure market.
The chipmaker introduced its AI200 and AI250 accelerator platforms, purpose-built for high-performance AI inference at scale. The AI200 is slated for release next year, followed by the AI250 in 2027. Both are designed to power generative AI workloads efficiently, offering high performance per dollar and per watt — a key metric in today’s energy-conscious data center industry.
The AI200 boasts 768 GB of LPDDR memory per card, delivering exceptional scale and flexibility for large language and multimodal models while keeping total cost of ownership low.
Meanwhile, the AI250 introduces a near-memory computing architecture that Qualcomm says will deliver more than 10 times higher effective memory bandwidth and dramatically lower power consumption—setting a new benchmark for AI inference efficiency.
Both systems support direct liquid cooling, PCIe scale-up, and Ethernet scale-out, ensuring optimal thermal performance and scalability at the rack level, with total power consumption of 160 kW.
As companies race to equip data centers for the AI era, Qualcomm’s move signals its determination to compete with Nvidia, AMD, and Intel in one of tech’s fastest-growing and most lucrative arenas.
