Qualcomm secures Meta CPU deal, targets $15B data center revenue

Qualcomm has positioned itself as a major player in AI infrastructure, moving beyond its traditional focus on edge devices. During its Investor Day event on June 25, 2026, the company revealed a series of strategic announcements, including a multigenerational CPU agreement with Meta and a new AI data center platform designed to address the demands of agentic AI workloads.

Qualcomm’s expansion into data center infrastructure reflects its ambition to diversify revenue streams. The company now expects data center revenue to exceed $15 billion annually by fiscal 2029, alongside significant growth in automotive, IoT, and industrial sectors. Handsets, once the core of Qualcomm’s business, are projected to account for roughly one-third of chip revenue under this forecast.

Meta deal and hyperscale partnerships

The multigenerational CPU agreement with Meta marks a validation of Qualcomm’s data center strategy. Meta CEO Mark Zuckerberg framed the partnership as critical to the company’s goal of delivering "personal superintelligence to everyone in the world." However, Meta emphasized a flexible approach, combining Qualcomm’s hardware with its own in-house MTIA silicon program. The company did not disclose deployment timelines, processor specifications, or specific workloads for the Qualcomm CPUs.

Beyond Meta, Qualcomm confirmed it has secured two major hyperscale customers, one of which remains unnamed. The deals are expected to generate at least $1 billion in revenue within a year, with initial shipments beginning by the end of 2026. Microsoft CEO Satya Nadella endorsed Qualcomm’s High-Bandwidth Compute (HBC) architecture during the event, though Microsoft did not announce a commercial deployment.

A broader platform for AI infrastructure

Qualcomm’s strategy extends beyond CPUs, encompassing AI accelerators, networking, custom silicon, and an open software stack. The company acquired Modular to strengthen its software capabilities, aiming to provide efficient AI workload execution across heterogeneous architectures. Tony Pialis, Qualcomm’s executive vice president and general manager of data center, argued that traditional server architectures cannot scale to meet the demands of agentic AI, necessitating a paradigm shift.

Central to this shift is High-Bandwidth Compute (HBC), a memory architecture designed to reduce bottlenecks during AI inference by combining SRAM-class performance with HBM-class capacity. Qualcomm did not release performance data or technical details, but analysts suggest HBC could improve inference efficiency, particularly in disaggregated AI infrastructure where data movement is a critical challenge.

Matt Kimball, vice president and principal analyst at Moor Insights & Strategy, noted that Qualcomm’s approach differentiates it from competitors by addressing multiple layers of AI infrastructure. "The CPU/xPU opportunity is huge," he said. "I think there is enough opportunity to support these players – and maybe a couple more."

Revenue and market implications

Qualcomm’s data center ambitions are underpinned by its financial projections. The company expects data center revenue to surpass $15 billion annually by fiscal 2029, with automotive, IoT, and industrial sectors contributing additional growth. Kimball described the target as ambitious but achievable, given the economics of hyperscale customers. "A relatively small number of large customer wins can translate into billions of dollars of annual revenue very quickly," he said.

The Meta deal, while not a market-shifting event on its own, provides Qualcomm with both revenue and credibility. Kimball emphasized that the partnership expands the conversation around Qualcomm’s role in the server CPU market, making it easier to attract additional cloud customers. However, the company’s long-term success will depend on its ability to deliver on its technical promises, particularly in memory architecture and software integration.