PrismML brings its tiny LLMs to Qualcomm-powered smart glasses
By Pradhyuman,
AI startup PrismML created a version of its language model for smart glasses running on Qualcomm Snapdragon chips. Caltech researchers founded the company, and UC Berkeley professor Ion Stoica serves as an advisor. At Qualcomm's Snapdragon Summit, the chipmaker showed PrismML's 1-bit Bonsai model running locally on the Snapdragon AR1 Gen 1 Platform.
The 2-billion-parameter model processes both vision and language. People wearing the glasses can ask questions about what they see in real time. PrismML stated that its system shrinks larger models by four times while keeping performance on standard benchmarks.
PrismML builds open-weight AI that runs on physical devices to use existing hardware power instead of relying on external AI providers. Device makers have not announced any commercial smart glasses using the PrismML model yet.
Maintained by Pradhyuman.
Filed under: Models & Research