Welcome to the TorchPipe Documentation!

TorchPipe is a multi-instance pipeline parallel library that provides a seamless integration between lower-level acceleration libraries (such as TensorRT and OpenCV) and RPC frameworks. It achieves strict decoupling from these libraries, offering a thread-safe function interface for the PyTorch frontend at a higher level. At the same time, it enables users to extend the backend capabilities at a lower level. TorchPipe's design ensures high service throughput while meeting latency requirements.