KServe flow: from an InferenceService to the first prediction

Scroll or pinch to zoom, drag to move, double-click to zoom in 100% SVG PNG Back to the post

KServe flow in five steps: the user applies an InferenceService; the controller creates the predictor pods with a matching serving runtime; the storage initializer downloads the model files from the model store; a client sends an inference request to the gateway; the gateway routes it to the predictor, which returns the prediction.