PrismML has deployed its lightweight language models on Qualcomm-powered smart glasses, enabling on-device AI processing without relying on cloud infrastructure.
The integration allows smart glasses equipped with Qualcomm processors to run language models locally, reducing latency and dependency on external servers. PrismML's approach focuses on optimized models designed for resource-constrained hardware.
The move aligns with PrismML's broader mission to develop open-weight AI that maximizes existing device computing capabilities. By shifting processing to the edge, the solution aims to improve performance efficiency while maintaining user privacy.
Qualcomm's mobile and wearable processors provide the computational foundation for these deployments. The partnership demonstrates growing momentum in bringing generative AI capabilities to consumer hardware beyond smartphones.
On-device LLMs remain a key priority for tech companies seeking to reduce server costs and improve user experience. Smart glasses represent an emerging application category where local processing offers particular advantages for real-time interactions.
An AI agent called Instinct saved one user $550 while handling restaurant bookings and detecting a phishing scam, but also wasted $64 and presents potential security concerns.
Google's Gemini 3.8 Live update introduces Live Avatar, an animated AI persona that responds to users with synchronized lip-syncing and facial expressions. The feature is currently limited to Gemini Enterprise customers.
Meta's consumer AI agent Muse has rocketed to the top of the App Store with 600,000 daily active US users, but the platform shares notable similarities with OpenClaw, the original AI agent framework.
Oracle faces mounting costs on a New Mexico data center project amid regulatory hurdles and local pushback. Meanwhile, Meta launches a palm-sized device for its Muse AI assistant, and President Trump meets with China's Xi Jinping to discuss artificial intelligence.