docker run -d --name hf-server \
  -p 8080:8080 \
  -e MODEL_NAME=bert-base-uncased \
  docker.io/kserve/huggingfaceserver:latest

kubectl apply -f - <<EOF
apiVersion: serving.kserve.io/v1beta1
kind: InferenceService
metadata:
  name: bert
spec:
  predictor:
    model:
      modelFormat:
        name: huggingface
      storageUri: gs://models/bert
      runtime: kserve-huggingface
EOF

apiVersion: apps/v1
kind: Deployment
metadata:
  name: hf-server
spec:
  replicas: 1
  selector:
    matchLabels:
      app: hf-server
  template:
    metadata:
      labels:
        app: hf-server
    spec:
      containers:
        - name: server
          image: docker.io/kserve/huggingfaceserver:0.13.0
          ports:
            - containerPort: 8080
          env:
            - name: MODEL_NAME
              value: bert-base-uncased
---
apiVersion: v1
kind: Service
metadata:
  name: hf-server
spec:
  selector:
    app: hf-server
  ports:
    - port: 8080
      targetPort: 8080