My pipeline is. Github self-hosted, running on the inference server with many Nvidia GPU's. Pretty easy to setup.