PrismML has brought its tiny language models to smart glasses running on Qualcomm’s Snapdragon chips, showcasing the development at the Snapdragon Summit on Wednesday. The AI lab, founded by Caltech researchers and advised by UC Berkeley’s Ion Stoica, demonstrated its 1-bit Bonsai LLM running locally on AI smart glasses built on the Snapdragon AR1 Gen 1 Platform.
The smart glasses version is a 2-billion-parameter model specifically tuned for vision and language capabilities, allowing wearers to ask questions about what they are looking at in real-time. PrismML shrinks larger models by 4x while retaining almost all of their performance on standard benchmarks, advancing the startup's goal of delivering open-weight on-device AI.
While releasing a model for Qualcomm’s Snapdragon chip is a significant step toward providing an alternative to proprietary cloud-based AI labs, no commercial smart glasses running PrismML have been announced yet. This on-device integration matters because it reduces reliance on cloud compute and offers enhanced privacy for wearable AI users.