XNews

APXInf open-source engine cuts embodied model latency 10.7x

APXInf open-source engine cuts embodied model latency 10.7x

APXInf, an open-source edge inference engine for embodied models, reduces reaction time from 278ms to under 26ms on Jetson Thor (10.7x faster). It supports PI 0.5 and WALL-OSS models, runs on RTX 4090 and Jetson boards, and is part of the RLinf training-to-deployment ecosystem.

Source posts