https://hackershare.dev/en/bookmarks/849625
Speed up model inference with Vertex AI Predictions’ optimized TensorFlow runtime