https://ops.docs.72602.space/kubernetes/serverless/kserve/serving/predictive/first_pytorch_infer/index.html Mnist Inference More Information about mnist service can be found 🔗link create a namespace kubectl create namespace kserve-test deploy a sample iris service kubectl apply -n kserve-test -f - <<EOF apiVersion: "serving.kserve.io/v1beta1" kind: "InferenceService" metadata: name: "first-torchserve" namespace: kserve-test spec: predictor: model: modelFormat: name: pytorch storageUri: gs://kfserving-examples/models/torchserve/image_classifier/v1 resources: limits: memory: 4Gi EOF Check InferenceService status kubectl -n kserve-test get inferenceservices first-torchserve Expectd Output kubectl -n kserve-test get pod #NAME READY STATUS RESTARTS AGE #first-torchserve-predictor-00001-deplo... 2/2 Running 0 25s kubectl -n kserve-test get inferenceservices first-torchserve #NAME URL READY PREV LATEST PREVROLLEDOUTREVISION LATESTREADYREVISION AGE #kserve-test first-torchserve http://first-torchserve.kserve-test.example.com True 100 first-torchserve-predictor-00001 2m59s After all pods are ready, you can access the service by using the following command Hugo en