Qualcomm has launched a new generation of Snapdragon smartphone chips designed to move more advanced AI processing directly onto mobile devices. The shift to a 2nm process and support for large AI models comes as the technology industry increasingly moves AI workloads from cloud infrastructure toward devices, where speed, privacy and power efficiency are becoming increasingly important.

Qualcomm Brings 2nm Technology to Flagship Smartphones

Qualcomm introduced the Snapdragon 8 Elite Extreme Gen 6 and Snapdragon 8 Elite Gen 6 at its Snapdragon Summit, expanding its premium mobile portfolio. Both platforms are built on an advanced 2nm process and combine Qualcomm’s Oryon CPU, redesigned Adreno GPU and next-generation Hexagon NPU to handle more demanding AI workloads.

The move to 2nm is significant because smartphone manufacturers need greater computing performance without a proportional increase in power consumption. Qualcomm is positioning the new platforms for agentic AI applications that can understand context, plan tasks and act on behalf of users. This could make smartphones more capable AI computing platforms rather than primarily interfaces for cloud-based services.

30 Billion Parameters Move On-Device

The Snapdragon 8 Elite Extreme Gen 6 is designed to support Mixture-of-Experts models with 30 billion parameters. Qualcomm’s new Hexagon NPU architecture includes a transformer-focused Element Accelerator and 50% more shared memory, allowing more model data to remain close to the processor while reducing trips to external memory.

The architecture is particularly relevant to larger AI models because a 30-billion-parameter Mixture-of-Experts model does not need to activate every parameter for each token. Qualcomm says such a model can keep tens of billions of parameters available while activating about 3 billion routed parameters for each token-generation step. This approach could support more responsive on-device AI while reducing the amount of information that needs to be sent to cloud servers.

AI Fits Into Qualcomm’s Broader Growth Strategy

The smartphone launch also fits into Qualcomm’s broader effort to diversify beyond handsets. In the third quarter of fiscal 2026, the company reported revenue of $9.9 billion, while combined QCT automotive and IoT revenue increased 28% year over year.

At its June 2026 Investor Day, Qualcomm raised its fiscal 2029 target for non-handset revenue to $40 billion, nearly double its previous target. The company is also targeting more than $15 billion in data-center revenue by fiscal 2029. Qualcomm expects handsets to represent approximately one-third of QCT revenue by that year, highlighting the importance of expanding its position across other computing markets.

The next test will be commercial adoption. Qualcomm will need to demonstrate that advanced on-device AI translates into meaningful demand from smartphone manufacturers and consumers. Investors will be watching the rollout of devices using the new platforms, real-world AI performance, power efficiency and adoption of agentic applications. At the same time, competition across mobile processors and AI platforms will help determine whether local AI becomes a meaningful growth driver for QCOM and a durable shift in smartphone computing.