PrismML is bringing its tiny large language models to smart glasses powered by Qualcomm chips. This move is part of a broader push for open-weight AI that runs directly on devices rather than relying on the cloud. By running models locally, the technology aims to maximise the computing power already available in the hardware.

The focus on on-device processing is a key part of PrismML's strategy. The company believes that using the power built into devices is a more efficient way to deliver AI capabilities. This approach is particularly relevant for wearable tech, where battery life and connectivity are critical factors.