PrismML's 1-Bit Bonsai LLM Debuts on Snapdragon AR1 Gen 1 Platform
PrismML, an AI lab founded by Caltech researchers and advised by UC Berkeley's Ion Stoica, has developed a version of its tiny language models specifically for smart glasses powered by Qualcomm's Snapdragon chips.
At Qualcomm's Snapdragon Summit on Wednesday, the chipmaker showcased PrismML's 1-bit Bonsai LLM, which runs locally on AI smart glasses built on the Snapdragon AR1 Gen 1 Platform.
Shrinking Models Without Sacrificing Performance
As TechCrunch previously reported, PrismML's key innovation is its ability to shrink larger models substantially—in this case, by a factor of four—while retaining nearly all of their performance on standard benchmarks. The smart glasses version is a 2-billion-parameter model optimized for both vision and language, enabling wearers to ask questions about what they are seeing in real time.
A Broader Vision for On-Device AI
PrismML's larger goal is to advance open-weight AI that runs directly on devices, making better use of the computing power they already possess. The startup positions this approach as an alternative to relying on the privacy assurances of proprietary AI labs and their relentless demand for more compute resources.
Releasing a model for Qualcomm's chip represents a step toward that vision, though no smart glasses running PrismML have been announced yet.