Prabhu RamQualcomm delivered a solid Q2 FY26, with revenue of $10.6 billion landing at the high end of guidance. More importantly, the quarter reinforces a structural shift in the company’s growth narrative—from reliance on scale in handsets to a more diversified, mix-driven model.
What’s Driving Growth
QCT revenue reached $9.1 billion, with Automotive emerging as the standout at $1.3 billion, up 38% year-on-year and marking a record quarter. Annualized automotive revenue has now crossed $5 billion, with visibility to exit FY26 above a $6 billion run rate—clear evidence that prior design wins are translating into realized revenue.
IoT remained steady at $1.7 billion, growing 9% year-on-year and continuing to provide diversification. Handsets, at $6.0 billion, were softer due to memory-driven supply constraints and cautious inventory behavior, particularly in China. However, premium-tier exposure remains resilient, supported by a greater than 70% Snapdragon share at Samsung.
Why It Matters
The key inflection is diversification reaching meaningful scale. Automotive and IoT combined grew 20% year-on-year, helping offset cyclicality in handsets. Automotive, in particular, is transitioning from a pipeline narrative to a revenue engine, driven by ADAS adoption, increasing compute requirements, and rising content per vehicle.
At the same time, agentic AI workloads are reshaping compute architectures, requiring continuous orchestration across CPU, NPU, and connectivity. Qualcomm’s strengths across these domains position it effectively across both edge and cloud environments. This extends into the data center, where the company is ramping custom silicon with a leading hyperscaler, with initial shipments expected in December.
The Big Picture
Near-term handset pressures remain cyclical, with a trough likely in Q3. Structurally, however, Qualcomm is reducing dependence on smartphones while building multi-engine growth across automotive, IoT, PCs, and data center.
An underappreciated advantage is its power-efficient architecture, optimized for inference at scale rather than training. As hyperscaler capex evolves, this positioning could become increasingly strategic.
