Qualcomm Powers AI Wearables With a 3nm “Wrist Plus” Chip

Qualcomm just unveiled an AI-first chip for wearables that fits on your wrist and beyond, signaling a shift toward more capable on-device intelligence.

Qualcomm announced the Snapdragon Wear Elite, a 3nm “wrist plus” processor designed to sit alongside the existing W5 Plus rather than replace it. The company framed the Elite as a way for gadget makers to build AI wearables that aren’t limited to traditional watches—think pendants, pins, and even display-free smart glasses. The hardware package includes both an on-device neural processing unit (NPU) and a Hexagon NPU, underscoring Qualcomm’s push to push heavier AI workloads out of the cloud and closer to the user.

The move comes at a time when wearables are increasingly expected to do more without draining battery life. A 3nm process is not just about raw speed; it promises meaningful power efficiency that matters for always-on features—heart-rate patterns, activity detection, ambient sensing, and on-device voice tasks. Qualcomm’s stated aim with the Wear Elite is to give creators a “wrist plus” platform—a term it used to describe devices that extend beyond a traditional wristwatch form factor. In practice, that means smarter jewelry-like wearables and discreet display-free devices that can still offer responsive AI-powered experiences.

Wearables makers have long wrestled with the tradeoffs between on-device processing and cloud-based AI. On-device inference can improve privacy and reduce latency, but it also squeezes hardware budgets in tiny shells and demands lower energy per operation. The Snapdragon Wear Elite’s 3nm node, and its dedicated NPUs, are meant to tilt that balance toward more sophisticated on-device AI while preserving battery life and thermal constraints. In plain terms: you should expect smarter health analytics, faster on-device assistants, and more context-aware features without sprinting toward a power brick.

Industry watchers will be watching which devices surface first with the Elite and how developers leverage the new capabilities. The move also hints at a broader pattern: AI is moving from the phone and earbud into the political and fashion-forward corners of the wearables market. Pendants and pins aren’t just fashion statements; with an on-chip AI brain, they could become more capable health monitors, EMF sensors, or discreet augmented-reality surfaces embedded into everyday objects. The anticipation is that a richer AI toolbox will unlock new form factors—earliest examples may arrive in the second half of the year as OEMs begin shipping prototypes and developer kits.

Two practitioner angles to watch:

Power budgets and thermal behavior in real-world wearables. Even with a smaller 3nm chip, sustained AI tasks—like continuous voice or pattern recognition—will test battery handling. Expect iterative cooling, tighter ML models, and possibly tiered inference that prioritizes the most critical tasks first.

The developer ecosystem and device roadmap. Elite’s value hinges on a robust SDK and partner roadmap. If creators can port models easily and exploit the eNPU and Hexagon NPU for common AI routines (voice, vision, biosignals) with predictable power use, we’ll see a faster wave of AI-enabled jewelry and glasses.

Qualcomm’s announcement signals more than a new chip. It marks a deliberate bet that AI-enabled wearables will no longer be tethered to bulky, screen-first devices. With the Wear Elite, the company is inviting a new wave of lightweight, context-aware wearables that quietly do more—without begging for a power outlet.

Sources

Qualcomm’s new chip is geared toward wearable AI gadgets