Qualcomm Investor Day 2026: Dragonfly CPU, $3.9B Modular Acquisition, and Meta as First Data Center Customer

Qualcomm is positioning itself as the "inference-first" player, arguing that the shift from training-centric to inference-centric computing creates a new architectural opportunity that incumbent x86 and GPU-only designs are less suited for .

Meta Platforms: First Named Data Center Customer

Qualcomm announced a strategic multi-generation supply agreement with Meta Platforms, which will be the first named customer for the Dragonfly C1000 . Under the deal, Qualcomm will supply Dragonfly C1000 CPUs to power Meta's next-generation server fleet, with production scheduled to begin in the second half of 2028 . Financial terms, including volume commitments or pricing benchmarks, were not disclosed . This was widely seen as the strongest validation yet of Qualcomm's data center ambitions .

Acquisition of Modular Inc. for $3.9 Billion

Qualcomm entered into a definitive agreement to acquire Modular Inc., an AI infrastructure software startup, for approximately $3.9 billion in stock . The deal, signed on June 21, 2026, will see Qualcomm issue 19.2 million shares to Modular's owners and bring the startup's entire 150-person team into the company . Modular's AI-native software platform is designed to let AI models run across different chips without hardware-specific recoding, which Qualcomm plans to use to strengthen its inference software stack . The deal is expected to close in the second half of 2026 . Modular's technology is viewed as a direct competitive angle against Nvidia's CUDA ecosystem . Modular had raised $250 million just nine months prior at a valuation of $1.6 billion .

Oryon-Based Data Center Strategy and Revenue Targets

Qualcomm outlined a comprehensive data center AI infrastructure strategy built around its custom Oryon cores . The company set a data center revenue target of more than $15 billion by fiscal 2029 . On the broader corporate level, Qualcomm raised its fiscal 2029 non-handset revenue target to $40 billion, nearly doubling its previous estimate of $22 billion . The automotive division alone is targeting $10 billion in revenue by fiscal 2029 . IoT revenue is targeted at more than $14 billion, with handsets expected to represent approximately one-third of QCT revenues by fiscal 2029 .

Note on the $3 billion by 2027 figure: Several pre-Investor Day analyst notes from J.P. Morgan projected Qualcomm would guide for data center revenue above $3 billion in fiscal 2027 . The official event materials published by Qualcomm set a $15 billion data center target by fiscal 2029, but did not explicitly reiterate a $3 billion FY2027 figure in widely reported headlines. The $3 billion figure was an analyst expectation that appears subsumed by the larger FY2029 target.

Competitive Landscape: How Qualcomm Stacks Up Against Nvidia, AMD, Intel, and Arm

Qualcomm's Investor Day framed its data center push as a challenge to the established order, entering a sector largely led by Nvidia .

• Nvidia: The Modular acquisition and Dragonfly inference accelerator are explicitly aimed at breaking Nvidia's dominance in AI infrastructure, offering an open-ecosystem alternative to Nvidia's proprietary CUDA platform . Nvidia CEO Jensen Huang had publicly praised Qualcomm ahead of the event, signaling the competitive tension .
• Arm: Qualcomm's Oryon cores are Arm-compatible, positioning the company within the broader Arm server ecosystem competing against x86 incumbents . Qualcomm and Arm remain in an ongoing legal and competitive dynamic over architecture licensing.
• Intel and AMD: Qualcomm enters a data center CPU market dominated by Intel (Xeon) and AMD (EPYC) for general-purpose servers, but Qualcomm's pitch focuses specifically on the inference-heavy, agentic AI workload segment rather than general-purpose server replacement . Analysts from GF Securities have named AMD, Intel, and Qualcomm as beneficiaries of an upcoming server CPU upgrade cycle tied to AI inference .
• Key differentiation: Qualcomm is positioning itself as the "inference-first" player optimized for agentic AI workloads, arguing that the shift from training-centric to inference-centric computing creates a new architectural opportunity that incumbent x86 and GPU-only designs are less suited for .

Market Reaction

Qualcomm's stock surged 15% in after-hours trading following the event, reflecting investor enthusiasm for the scale of the data center ambition and the Meta win . However, during the regular trading session on June 24, shares traded at approximately $193.61, down 5.15%, as investors weighed the short-term costs of the approximately $4 billion Modular acquisition against data center revenue not expected until the second half of 2028 .