service-catalog/charts/kubeflow/contrib/kserve/kserve/kserve-cluster-resources.yaml

apiVersion: serving.kserve.io/v1alpha1
kind: ClusterServingRuntime
metadata:
  name: kserve-huggingfaceserver
spec:
  annotations:
    prometheus.kserve.io/path: /metrics
    prometheus.kserve.io/port: "8080"
  containers:
  - args:
    - --model_name={{.Name}}
    image: kserve/huggingfaceserver:v0.13.0
    name: kserve-container
    resources:
      limits:
        cpu: "1"
        memory: 2Gi
      requests:
        cpu: "1"
        memory: 2Gi
  protocolVersions:
  - v2
  - v1
  supportedModelFormats:
  - autoSelect: true
    name: huggingface
    priority: 1
    version: "1"
---
apiVersion: serving.kserve.io/v1alpha1
kind: ClusterServingRuntime
metadata:
  name: kserve-lgbserver
spec:
  annotations:
    prometheus.kserve.io/path: /metrics
    prometheus.kserve.io/port: "8080"
  containers:
  - args:
    - --model_name={{.Name}}
    - --model_dir=/mnt/models
    - --http_port=8080
    - --nthread=1
    image: kserve/lgbserver:v0.13.0
    name: kserve-container
    resources:
      limits:
        cpu: "1"
        memory: 2Gi
      requests:
        cpu: "1"
        memory: 2Gi
  protocolVersions:
  - v1
  - v2
  supportedModelFormats:
  - autoSelect: true
    name: lightgbm
    priority: 1
    version: "3"
---
apiVersion: serving.kserve.io/v1alpha1
kind: ClusterServingRuntime
metadata:
  name: kserve-mlserver
spec:
  annotations:
    prometheus.kserve.io/path: /metrics
    prometheus.kserve.io/port: "8080"
  containers:
  - env:
    - name: MLSERVER_MODEL_IMPLEMENTATION
      value: '{{.Labels.modelClass}}'
    - name: MLSERVER_HTTP_PORT
      value: "8080"
    - name: MLSERVER_GRPC_PORT
      value: "9000"
    - name: MODELS_DIR
      value: /mnt/models
    image: docker.io/seldonio/mlserver:1.3.2
    name: kserve-container
    resources:
      limits:
        cpu: "1"
        memory: 2Gi
      requests:
        cpu: "1"
        memory: 2Gi
  protocolVersions:
  - v2
  supportedModelFormats:
  - autoSelect: true
    name: sklearn
    priority: 2
    version: "0"
  - autoSelect: true
    name: sklearn
    priority: 2
    version: "1"
  - autoSelect: true
    name: xgboost
    priority: 2
    version: "1"
  - autoSelect: true
    name: lightgbm
    priority: 2
    version: "3"
  - autoSelect: true
    name: mlflow
    priority: 1
    version: "1"
---
apiVersion: serving.kserve.io/v1alpha1
kind: ClusterServingRuntime
metadata:
  name: kserve-paddleserver
spec:
  annotations:
    prometheus.kserve.io/path: /metrics
    prometheus.kserve.io/port: "8080"
  containers:
  - args:
    - --model_name={{.Name}}
    - --model_dir=/mnt/models
    - --http_port=8080
    image: kserve/paddleserver:v0.13.0
    name: kserve-container
    resources:
      limits:
        cpu: "1"
        memory: 2Gi
      requests:
        cpu: "1"
        memory: 2Gi
  protocolVersions:
  - v1
  - v2
  supportedModelFormats:
  - autoSelect: true
    name: paddle
    priority: 1
    version: "2"
---
apiVersion: serving.kserve.io/v1alpha1
kind: ClusterServingRuntime
metadata:
  name: kserve-pmmlserver
spec:
  annotations:
    prometheus.kserve.io/path: /metrics
    prometheus.kserve.io/port: "8080"
  containers:
  - args:
    - --model_name={{.Name}}
    - --model_dir=/mnt/models
    - --http_port=8080
    image: kserve/pmmlserver:v0.13.0
    name: kserve-container
    resources:
      limits:
        cpu: "1"
        memory: 2Gi
      requests:
        cpu: "1"
        memory: 2Gi
  protocolVersions:
  - v1
  - v2
  supportedModelFormats:
  - autoSelect: true
    name: pmml
    priority: 1
    version: "3"
  - autoSelect: true
    name: pmml
    priority: 1
    version: "4"
---
apiVersion: serving.kserve.io/v1alpha1
kind: ClusterServingRuntime
metadata:
  name: kserve-sklearnserver
spec:
  annotations:
    prometheus.kserve.io/path: /metrics
    prometheus.kserve.io/port: "8080"
  containers:
  - args:
    - --model_name={{.Name}}
    - --model_dir=/mnt/models
    - --http_port=8080
    image: kserve/sklearnserver:v0.13.0
    name: kserve-container
    resources:
      limits:
        cpu: "1"
        memory: 2Gi
      requests:
        cpu: "1"
        memory: 2Gi
  protocolVersions:
  - v1
  - v2
  supportedModelFormats:
  - autoSelect: true
    name: sklearn
    priority: 1
    version: "1"
---
apiVersion: serving.kserve.io/v1alpha1
kind: ClusterServingRuntime
metadata:
  name: kserve-tensorflow-serving
spec:
  annotations:
    prometheus.kserve.io/path: /metrics
    prometheus.kserve.io/port: "8080"
  containers:
  - args:
    - --model_name={{.Name}}
    - --port=9000
    - --rest_api_port=8080
    - --model_base_path=/mnt/models
    - --rest_api_timeout_in_ms=60000
    command:
    - /usr/bin/tensorflow_model_server
    image: tensorflow/serving:2.6.2
    name: kserve-container
    resources:
      limits:
        cpu: "1"
        memory: 2Gi
      requests:
        cpu: "1"
        memory: 2Gi
  protocolVersions:
  - v1
  - grpc-v1
  supportedModelFormats:
  - autoSelect: true
    name: tensorflow
    priority: 2
    version: "1"
  - autoSelect: true
    name: tensorflow
    priority: 2
    version: "2"
---
apiVersion: serving.kserve.io/v1alpha1
kind: ClusterServingRuntime
metadata:
  name: kserve-torchserve
spec:
  annotations:
    prometheus.kserve.io/path: /metrics
    prometheus.kserve.io/port: "8082"
  containers:
  - args:
    - torchserve
    - --start
    - --model-store=/mnt/models/model-store
    - --ts-config=/mnt/models/config/config.properties
    env:
    - name: TS_SERVICE_ENVELOPE
      value: '{{.Labels.serviceEnvelope}}'
    image: pytorch/torchserve-kfs:0.9.0
    name: kserve-container
    resources:
      limits:
        cpu: "1"
        memory: 2Gi
      requests:
        cpu: "1"
        memory: 2Gi
  protocolVersions:
  - v1
  - v2
  - grpc-v2
  supportedModelFormats:
  - autoSelect: true
    name: pytorch
    priority: 2
    version: "1"
---
apiVersion: serving.kserve.io/v1alpha1
kind: ClusterServingRuntime
metadata:
  name: kserve-tritonserver
spec:
  annotations:
    prometheus.kserve.io/path: /metrics
    prometheus.kserve.io/port: "8002"
  containers:
  - args:
    - tritonserver
    - --model-store=/mnt/models
    - --grpc-port=9000
    - --http-port=8080
    - --allow-grpc=true
    - --allow-http=true
    image: nvcr.io/nvidia/tritonserver:23.05-py3
    name: kserve-container
    resources:
      limits:
        cpu: "1"
        memory: 2Gi
      requests:
        cpu: "1"
        memory: 2Gi
  protocolVersions:
  - v2
  - grpc-v2
  supportedModelFormats:
  - autoSelect: true
    name: tensorrt
    priority: 1
    version: "8"
  - autoSelect: true
    name: tensorflow
    priority: 1
    version: "1"
  - autoSelect: true
    name: tensorflow
    priority: 1
    version: "2"
  - autoSelect: true
    name: onnx
    priority: 1
    version: "1"
  - name: pytorch
    version: "1"
  - autoSelect: true
    name: triton
    priority: 1
    version: "2"
---
apiVersion: serving.kserve.io/v1alpha1
kind: ClusterServingRuntime
metadata:
  name: kserve-xgbserver
spec:
  annotations:
    prometheus.kserve.io/path: /metrics
    prometheus.kserve.io/port: "8080"
  containers:
  - args:
    - --model_name={{.Name}}
    - --model_dir=/mnt/models
    - --http_port=8080
    - --nthread=1
    image: kserve/xgbserver:v0.13.0
    name: kserve-container
    resources:
      limits:
        cpu: "1"
        memory: 2Gi
      requests:
        cpu: "1"
        memory: 2Gi
  protocolVersions:
  - v1
  - v2
  supportedModelFormats:
  - autoSelect: true
    name: xgboost
    priority: 1
    version: "1"
---
apiVersion: serving.kserve.io/v1alpha1
kind: ClusterStorageContainer
metadata:
  name: default
spec:
  container:
    image: kserve/storage-initializer:v0.13.0
    name: storage-initializer
    resources:
      limits:
        cpu: "1"
        memory: 1Gi
      requests:
        cpu: 100m
        memory: 100Mi
  supportedUriFormats:
  - prefix: gs://
  - prefix: s3://
  - prefix: hdfs://
  - prefix: webhdfs://
  - regex: https://(.+?).blob.core.windows.net/(.+)
  - regex: https://(.+?).file.core.windows.net/(.+)
  - regex: https?://(.+)/(.+)