Not exactly a single binary, but Flogo - https://github.com/TIBCOSoftware/flogo uses one binary + one .so for TF inferencing. See https://tibcosoftware.github.io/flogo/development/flows/tens... for some basic documentation.

We’re looking to grow this further with a model serving service so that its more controlled and better abstracted from processing logic - A/B, champion/challenger, etc. Core devs hang out on gitter in case you have qns or feedback.