TrashAI-API-CPU

This is the proper implementation of this API. It is only compatible with the Docker deployment. You can find out more about how to deploy it in the README.md file.

TrashAI-API-WebGL v1.1

Bare-bones Functionality, as stated in README.md

It is exponentially slower than the CPU version, but doesn't require Docker. Can be used for testing purposes to check the applicability of this model within your project, before committing to it.