A collection of libraries to optimise AI model performances

A batched offline inference oriented version of segment-anything

YOLOv6: a single-stage object detection framework dedicated to industrial applications.

TensorFlow Lite backend with ArmNN delegate support for Nvidia Triton

Proxy based Redis cluster solution supporting pipeline and scaling dynamically

A Multithreaded Fork of Redis

Magface Triton Inferece Server Using Tensorrt

Streaming from ffmpeg to HTML5 canvas.