servingEngineSpec:
  modelSpec:
  - name: "llama3"
    repository: "vllm/vllm-openai"
    tag: "latest"
    modelURL: "meta-llama/Llama-3.1-8B-Instruct"
    replicaCount: 1

    requestCPU: 10
    requestMemory: "16Gi"
    requestGPU: 1

    pvcStorage: "50Gi"

    vllmConfig:
      enableChunkedPrefill: false
      enablePrefixCaching: false
      maxModelLen: 24576
      dtype: "float16"
      extraArgs: ["--disable-log-requests", "--gpu-memory-utilization", "0.8"]
    hf_token: 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

routerSpec:
  # -- The docker image of the router. The following values are defaults:
  repository: "lmcache/lmstack-router"
  tag: "latest"
  enableRouter: true
  resources:
    requests:
      cpu: "2"
      memory: "8G"
    limits:
      cpu: "4"
      memory: "16G"
  ingress:
    enabled: true
    className: ""
    annotations:
      cert-manager.io/cluster-issuer: "selfsigned-issuer" 
      cert-manager.io/duration: 8760h  
      cert-manager.io/renew-before: 720h
    hosts:
      - host: vllm.example.com
        paths:
          - path: /
            pathType: Prefix
    tls:
     - secretName: vllm-tls-secret
       hosts:
         - vllm.example.com