An AI inference optimization framework. This projects integrates major model compression algorithms in an unified way, to enable developers use faster, smaller, more efficient and greener models with a minimal overhead.