A very thin python library providing async streaming inferencing to LLaMA.cpp's HTTP Server via the API endpoints e.g. /completion. While you could get up and running quickly using something like ...