LiteRT.js runs machine learning models locally with CPU, GPU and emerging NPU acceleration, potentially reducing server infrastructure, inference charges and data movement.
Some results have been hidden because they may be inaccessible to you
Show inaccessible results