Founded by Caltech researchers and advised by UC Berkeley’s Ion Stoica, PrismML has developed a specialized version of its compact language models specifically designed for smart glasses powered by Qualcomm’s Snapdragon processors.
At the Snapdragon Summit held on Wednesday, Qualcomm demonstrated PrismML’s 1-bit Bonsai Large Language Model (LLM), a capability that allows for local execution on AI smart glasses designed with the Snapdragon AR1 Gen 1 Platform.
As TechCrunch previously highlighted, PrismML’s primary innovation involves drastically reducing the size of larger models—specifically by 4x in this instance—while maintaining nearly identical performance on standard benchmarks. The smart glasses iteration serves as a 2-billion-parameter model optimized for vision and language tasks, enabling users to query their surroundings in real time.
PrismML’s broader vision centers on open-weight AI that operates directly on devices, effectively maximizing the computing power already available. The startup positions this approach as a viable alternative to relying on proprietary AI laboratories and their growing demand for additional computational resources.
Making this model compatible with Qualcomm’s chips represents a significant step toward this goal, although the startup has not yet announced any smart glasses products utilizing PrismML’s technology.
The Editorial Staff at AIChief is a team of professional content writers with extensive experience in AI and marketing. Founded in 2025, AIChief has quickly grown into the largest free AI resource hub in the industry.