underneathall/pinferencia
A Python library that turns machine learning models into inference servers with minimal code, providing REST APIs and a GUI.

Pinferencia is a model deployment and serving library designed to be the simplest way to put machine learning models online. It allows users to serve any model with just three extra lines of code, providing both a REST API endpoint and a web-based GUI for interacting with models. The library supports popular frameworks including PyTorch, TensorFlow, HuggingFace Transformers, and PaddlePaddle, handling tasks across computer vision and NLP domains.