Qualcomm unveiled one of its biggest updates in years for the wearables segment at Mobile World Congress 2026 in Barcelona: the Snapdragon Wear Elite. It’s a new platform designed specifically to bring artificial intelligence directly to the device, without constantly relying on the phone or cloud servers — and it’s already been adopted by Samsung in its recently launched Galaxy Watch Ultra2.

The First Wearable Chip With a Dedicated NPU

The platform’s biggest headline is that it’s the first in its category to include a dedicated NPU (neural processing unit), named Hexagon. This component can run AI models with up to 2 billion parameters directly on the device, eliminating steps that were previously unavoidable: sending audio or data to the phone, from there to the cloud, and then back to the watch.

Beyond this primary NPU, the chip also integrates a secondary eNPU, built for low-power, always-on AI tasks — a key consideration on a device where space and battery are both limited. Among the use cases Qualcomm highlights are activity logging with health recommendations, personalized training, text summarization, and real-time transcription and translation.

Performance: Up to Five Times Faster

On the architecture side, the Snapdragon Wear Elite is built on a 3-nanometer process and features five cores: one high-performance core at 2.1 GHz and four efficiency cores at 1.9 GHz. This configuration translates, according to Qualcomm, into CPU performance up to five times faster in single-thread processes and up to seven times faster graphics compared to the previous generation of wearable chips.

The company also promises 30% more battery life and considerably faster charging: 0% to 50% in just 10 minutes — a meaningful leap for devices that have historically been criticized for shorter battery life compared to the competition.

Hexa-Connectivity: 5G and Satellite Messaging

Qualcomm coined the term “hexa-connectivity” for the Snapdragon Wear Elite’s connection package, which includes 5G RedCap, Bluetooth 6.0, UWB, GNSS, and, for the first time in this segment, NB-NTN-based satellite messaging support. This combination opens the door to watches capable of staying connected even in areas without traditional cellular coverage — a function that until now was reserved almost exclusively for some high-end smartphones.

Beyond the Wrist: Pins, Pendants, and Personal Hubs

One of the most interesting aspects of this platform is that it’s not built exclusively for watches. Qualcomm envisions it as a foundation for a wider range of form factors: smart glasses, conversational pins, pendants, and personal hubs — categories that barely exist at scale today but could take off thanks to hardware like this. The platform supports Wear OS, Android, and Linux, giving it flexibility to adapt across different systems and manufacturers.

Qualcomm also mentions a more ambitious concept within this ecosystem: an AI assistant acting as a “digital proxy,” capable of performing tasks on the user’s behalf, leveraging the chip’s local processing power to act faster and without depending on a constant connection.

Privacy: The Advantage of On-Device Processing

Qualcomm emphasizes that processing AI locally, directly on the chip, offers an additional advantage beyond speed: privacy. By not relying on sending personal or biometric data to external servers for processing, the company argues this approach reduces exposure of sensitive user information — an increasingly important argument for devices that track health data around the clock.

Industry Backing

The new platform already has backing from key players in the sector, including Samsung, Motorola, and Google, all of which have publicly supported Snapdragon Wear Elite as the technological foundation for their upcoming wearables. In fact, Samsung already confirmed its adoption in the Galaxy Watch Ultra2, unveiled just months after Qualcomm’s announcement, suggesting fast adoption among major Wear OS watchmakers.

The first commercial products built on this chip are expected to keep arriving over the coming months, ushering in a new generation of wearables that, for the first time, won’t rely almost entirely on the smartphone to deliver advanced AI features.