values.yaml

## @section Global parameters
## Global Docker image parameters
## @param global.image.registry Global Docker image registry
## @param global.image.repository Global Docker image repository
## @param global.image.tag Global Docker image tag (immutable tags are recommended)
global:
  image:
    registry: quay.io
    repository: activeloopai/neohorizon-main
    tag: v2.5.2
  ## @param global.nodeSelector Global node selector for all pods
  nodeSelector: {}
  ## @param global.tolerations Global tolerations for all pods
  tolerations: []
  ## @param global.affinity Global affinity for all pods
  affinity: {}
  ## @section Global configuration parameters [Description of the parameters](https://github.com/activeloopai/activeloop-self-hosted-resources/blob/main/README.md#Parameters)
  ## @param global.config.postgres_database PostgreSQL database name (default: see `postgresql.auth.database`)
  ## @param global.config.postgres_host PostgreSQL host address (default: see `postgresql.primary.persistence.host`)
  ## @param global.config.postgres_password PostgreSQL password (default: see `postgresql.auth.password`)
  ## @param global.config.postgres_port PostgreSQL port (default: see `postgresql.primary.persistence.port`)
  ## @param global.config.postgres_user PostgreSQL username (default: see `postgresql.auth.username`)
  ## @param global.config.deeplake_creds Deep Lake credentials, for details refer to [README](https://github.com/activeloopai/activeloop-self-hosted-resources?tab=readme-ov-file#deeplake-storage-credentials)
  ## @param global.config.deeplake_root_dir Deep Lake storage root, like s3://my-bucket/my-dir
  ## @param global.config.rabbitmq_url RabbitMQ connection URL, default is generated from rabbitmq.auth.url
  ## @param global.config.al_api_token API key for api authentication
  ## @param global.config.gemini_api_key Gemini API key for gemini model
  ## @param global.config.openai_api_key OpenAI API key for openai model
  ## @param global.config.text_image__matrix_of_embeddings__ingestion_url Text image matrix of embeddings ingestion model URL
  ## @param global.config.text_image__matrix_of_embeddings__query_url Text image matrix of embeddings query model URL
  ## @param global.config.text_image__embedding__ingestion_url Text image embedding ingestion model URL
  ## @param global.config.text_image__embedding__query_url Text image embedding query model URL
  ## @param global.config.text__embedding__ingestion_url Text embedding ingestion model URL
  ## @param global.config.text__embedding__query_url Text embedding query model URL
  ## @param global.config.text__summary_embedding__query_url summary embedding query model URL
  ## @param global.config.text__summary_embedding__ingestion_url summary embedding ingestion model URL
  ## @param global.config.gotenberg_endpoint Gotenberg endpoint, disabled by default (needed for office files processing)
  ## @param global.config.chunkr_api_key Chunkr API key, disabled by default (needed for xlsx files processing)
  config:
    postgres_database: "neohorizon"
    postgres_password: "postgres"
    postgres_port: "5432"
    postgres_user: "postgres"
    postgres_host: ""
    rabbitmq_url: ""
    deeplake_creds: ""
    deeplake_root_dir: ""
    al_api_token: ""
    gemini_api_key: ""
    openai_api_key: ""
    text_image__matrix_of_embeddings__ingestion_url: ""
    text_image__matrix_of_embeddings__query_url: ""
    text_image__embedding__ingestion_url: ""
    text_image__embedding__query_url: ""
    text__embedding__ingestion_url: ""
    text__embedding__query_url: ""
    text__summary_embedding__query_url: ""
    text__summary_embedding__ingestion_url: ""
    gotenberg_endpoint: ""
    chunkr_api_key: ""
  ## @param global.useExistingSecret Use existing secret for the config, if set config will be ignored
  # -- secret will be referenced as envFrom, make sure keys are uppercase like POSTGRES_PASSWORD, etc.
  # -- make sure to set following values as well
  #   USE_DEEPLAKE: "true"
  #   USE_RABBITMQ: "true"
  #   USE_DEEPLAKE_AUTH: "false"
  #   ENVIRONMENT: "local"
  #   EMBEDDING_PROVIDER_TYPE: "TRITON"
  #   CORS_ALLOW_ORIGINS: "*"
  useExistingSecret: ""
## @section Ingress parameters
## @param ingress.enabled Enable ingress record generation for the application
## @param ingress.className IngressClass that will be used to implement the Ingress
## @param ingress.annotations Additional custom annotations for the ingress record
## @param ingress.labels Additional custom labels for the ingress record
## @param ingress.host Default host for the ingress record
## @param ingress.tls Enable TLS configuration for the hostname defined at ingress.host parameter
ingress:
  enabled: false
  className: ""
  annotations: {}
  labels: {}
  host: "api.example.com"
  tls: []
  # tls:
  #   - hosts:
  #       - "api.example.com"
  #     secretName: "neohorizon-tls"
## @section API Services parameters
apis:
  ## @param apis.lightweight_service.enabled Enable lightweight service deployment
  ## @param apis.lightweight_service.replicas Number of lightweight service replicas to deploy
  ## @param apis.lightweight_service.port Port for the lightweight service
  ## @param apis.lightweight_service.annotations Annotations for the lightweight service deployment
  ## @param apis.lightweight_service.strategy.type Deployment strategy type
  ## @param apis.lightweight_service.scaling Horizontal Pod Autoscaler configuration
  ## @param apis.lightweight_service.command Override default container command
  ## @param apis.lightweight_service.args Override default container args
  ## @param apis.lightweight_service.service.type Kubernetes service type
  ## @param apis.lightweight_service.service.port Service port
  ## @param apis.lightweight_service.env Environment variables for the lightweight service
  ## @param apis.lightweight_service.envFrom Environment variables from ConfigMap/Secret
  ## @skip apis.lightweight_service.resources.requests.cpu
  ## @skip apis.lightweight_service.resources.requests.memory
  ## @param apis.lightweight_service.serviceAccount.create Specifies whether a service account should be created
  ## @param apis.lightweight_service.serviceAccount.name The name of the service account to use
  ## @param apis.lightweight_service.serviceAccount.labels Additional labels for the service account
  ## @param apis.lightweight_service.serviceAccount.annotations Additional annotations for the service account
  ## @param apis.lightweight_service.enableHealthProbes Enable health probes for the lightweight service
  ## @param apis.lightweight_service.nodeSelector Node selector for lightweight service pods
  ## @param apis.lightweight_service.tolerations Tolerations for lightweight service pods
  ## @param apis.lightweight_service.affinity Affinity for lightweight service pods
  lightweight_service:
    enabled: true
    replicas: 1
    port: 8000
    annotations: {}
    strategy:
      type: RollingUpdate
    scaling: {}
    # minReplicas: 2
    # maxReplicas: 5
    # metrics:
    #   cpu: 70
    #   memory: 50
    command: []
    args: []
    service:
      type: ClusterIP
      port: 80
    env: {}
    envFrom: []
    resources:
      requests:
        cpu: "1"
        memory: "1Gi"
        #limits:
        #  cpu: "2"
        #  memory: "2Gi"
    serviceAccount:
      create: false
      name: ""
      labels: {}
      annotations: {}
    enableHealthProbes: true
    nodeSelector: {}
    tolerations: []
    affinity: {}
  ## @param apis.files_service.enabled Enable files service deployment
  ## @param apis.files_service.replicas Number of files service replicas to deploy
  ## @param apis.files_service.port Port for the files service
  ## @param apis.files_service.strategy.type Deployment strategy type
  ## @param apis.files_service.scaling Horizontal Pod Autoscaler configuration
  ## @param apis.files_service.image.repository Docker image repository for files service
  ## @param apis.files_service.command Override default container command
  ## @param apis.files_service.args Override default container args
  ## @param apis.files_service.service.type Kubernetes service type
  ## @param apis.files_service.service.port Service port
  ## @param apis.files_service.env Environment variables for the files service
  ## @param apis.files_service.envFrom Environment variables from ConfigMap/Secret
  ## @skip apis.files_service.resources.requests.cpu
  ## @skip apis.files_service.resources.requests.memory
  ## @param apis.files_service.serviceAccount.create Specifies whether a service account should be created
  ## @param apis.files_service.serviceAccount.name The name of the service account to use
  ## @param apis.files_service.serviceAccount.labels Additional labels for the service account
  ## @param apis.files_service.serviceAccount.annotations Additional annotations for the service account
  ## @param apis.files_service.enableHealthProbes Enable health probes for the files service
  ## @param apis.files_service.nodeSelector Node selector for files service pods
  ## @param apis.files_service.tolerations Tolerations for files service pods
  ## @param apis.files_service.affinity Affinity for files service pods
  files_service:
    enabled: true
    replicas: 1
    port: 8000
    strategy:
      type: RollingUpdate
    scaling: {}
    # minReplicas: 3
    # maxReplicas: 6
    # metrics:
    #   cpu: 80
    #   memory: 80
    image:
      repository: activeloopai/neohorizon-files
    command: []
    args: []
    service:
      type: ClusterIP
      port: 80
    env: {}
    envFrom: []
    resources:
      requests:
        cpu: "2"
        memory: 4Gi
        #limits:
        #  cpu: "3"
        #  memory: 8Gi
    serviceAccount:
      create: false
      name: ""
      labels: {}
      annotations: {}
    enableHealthProbes: true
    nodeSelector: {}
    tolerations: []
    affinity: {}
  ## @param apis.nango_webhook_service.enabled Enable nango webhook service deployment
  ## @param apis.nango_webhook_service.replicas Number of files service replicas to deploy
  ## @param apis.nango_webhook_service.port Port for the nango webhook service
  ## @param apis.nango_webhook_service.strategy.type Deployment strategy type
  ## @param apis.nango_webhook_service.scaling Horizontal Pod Autoscaler configuration
  ## @param apis.nango_webhook_service.image.repository Docker image repository for files service
  ## @param apis.nango_webhook_service.command Override default container command
  ## @param apis.nango_webhook_service.args Override default container args
  ## @param apis.nango_webhook_service.service.type Kubernetes service type
  ## @param apis.nango_webhook_service.service.port Service port
  ## @param apis.nango_webhook_service.env Environment variables for the files service
  ## @param apis.nango_webhook_service.envFrom Environment variables from ConfigMap/Secret
  ## @skip apis.nango_webhook_service.resources.requests.cpu
  ## @skip apis.nango_webhook_service.resources.requests.memory
  ## @param apis.nango_webhook_service.serviceAccount.create Specifies whether a service account should be created
  ## @param apis.nango_webhook_service.serviceAccount.name The name of the service account to use
  ## @param apis.nango_webhook_service.serviceAccount.labels Additional labels for the service account
  ## @param apis.nango_webhook_service.serviceAccount.annotations Additional annotations for the service account
  ## @param apis.nango_webhook_service.enableHealthProbes Enable health probes for the files service
  ## @param apis.nango_webhook_service.nodeSelector Node selector for files service pods
  ## @param apis.nango_webhook_service.tolerations Tolerations for files service pods
  ## @param apis.nango_webhook_service.affinity Affinity for files service pods
  nango_webhook_service:
    enabled: false
    replicas: 1
    port: 8080
    strategy:
      type: RollingUpdate
    scaling: {}
    # minReplicas: 3
    # maxReplicas: 6
    # metrics:
    #   cpu: 80
    #   memory: 80
    image:
      repository: activeloopai/neohorizon-nango
    command: []
    args: []
    service:
      type: ClusterIP
      port: 80
    env: {}
    envFrom: []
    resources:
      requests:
        cpu: "100m"
        memory: "256Mi"
        #limits:
        #  cpu: "3"
        #  memory: 8Gi
    serviceAccount:
      create: false
      name: ""
      labels: {}
      annotations: {}
    enableHealthProbes: true
    nodeSelector: {}
    tolerations: []
    affinity: {}
## @section Workers parameters
workers:
  ## @param workers.vector_search_worker.enabled Enable vector search worker deployment
  ## @param workers.vector_search_worker.replicas Number of vector search worker replicas to deploy
  ## @param workers.vector_search_worker.annotations Annotations for the vector search worker deployment
  ## @param workers.vector_search_worker.strategy.type Deployment strategy type
  ## @param workers.vector_search_worker.scaling Horizontal Pod Autoscaler configuration
  ## @param workers.vector_search_worker.env Environment variables for the vector search worker
  ## @param workers.vector_search_worker.envFrom Environment variables from ConfigMap/Secret
  ## @skip workers.vector_search_worker.resources.requests.cpu
  ## @skip workers.vector_search_worker.resources.requests.memory
  ## @param workers.vector_search_worker.serviceAccount.create Specifies whether a service account should be created
  ## @param workers.vector_search_worker.serviceAccount.name The name of the service account to use
  ## @param workers.vector_search_worker.serviceAccount.labels Additional labels for the service account
  ## @param workers.vector_search_worker.serviceAccount.annotations Additional annotations for the service account
  ## @param workers.vector_search_worker.enableHealthProbes Enable health probes for the vector search worker
  ## @param workers.vector_search_worker.nodeSelector Node selector for vector search worker pods
  ## @param workers.vector_search_worker.tolerations Tolerations for vector search worker pods
  ## @param workers.vector_search_worker.affinity Affinity for vector search worker pods
  vector_search_worker:
    enabled: true
    replicas: 1
    annotations: {}
    strategy:
      type: RollingUpdate
    scaling: {}
    # minReplicas: 1
    # maxReplicas: 5
    # metrics:
    #   cpu: 70
    #   memory: 70
    env: {}
    envFrom: []
    resources:
      requests:
        cpu: "12"
        memory: 24Gi
        #limits:
        #  cpu: "15"
        #  memory: 30Gi
    serviceAccount:
      create: false
      name: ""
      labels: {}
      annotations: {}
    enableHealthProbes: true
    nodeSelector: {}
    tolerations: []
    affinity: {}
  ## @param workers.file_processor_worker.enabled Enable file processor worker deployment
  ## @param workers.file_processor_worker.replicas Number of file processor worker replicas to deploy
  ## @param workers.file_processor_worker.annotations Annotations for the file processor worker deployment
  ## @param workers.file_processor_worker.strategy.type Deployment strategy type
  ## @param workers.file_processor_worker.scaling Horizontal Pod Autoscaler configuration
  ## @param workers.file_processor_worker.env Environment variables for the file processor worker
  ## @param workers.file_processor_worker.envFrom Environment variables from ConfigMap/Secret
  ## @skip workers.file_processor_worker.resources.requests.cpu
  ## @skip workers.file_processor_worker.resources.requests.memory
  ## @param workers.file_processor_worker.serviceAccount.create Specifies whether a service account should be created
  ## @param workers.file_processor_worker.serviceAccount.name The name of the service account to use
  ## @param workers.file_processor_worker.serviceAccount.labels Additional labels for the service account
  ## @param workers.file_processor_worker.serviceAccount.annotations Additional annotations for the service account
  ## @param workers.file_processor_worker.enableHealthProbes Enable health probes for the file processor worker
  ## @param workers.file_processor_worker.nodeSelector Node selector for file processor worker pods
  ## @param workers.file_processor_worker.tolerations Tolerations for file processor worker pods
  ## @param workers.file_processor_worker.affinity Affinity for file processor worker pods
  file_processor_worker:
    enabled: true
    replicas: 1
    annotations: {}
    strategy:
      type: RollingUpdate
    scaling: {}
    # minReplicas: 1
    # maxReplicas: 5
    # metrics:
    #   cpu: 70
    #   memory: 70
    env: {}
    envFrom: []
    resources:
      requests:
        cpu: "4"
        memory: 8Gi
        #limits:
        #  cpu: "8"
        #  memory: 15Gi
    serviceAccount:
      create: false
      name: ""
      labels: {}
      annotations: {}
    enableHealthProbes: true
    nodeSelector: {}
    tolerations: []
    affinity: {}
  ## @param workers.embedding_worker.enabled Enable embedding worker deployment
  ## @param workers.embedding_worker.replicas Number of embedding worker replicas to deploy
  ## @param workers.embedding_worker.annotations Annotations for the embedding worker deployment
  ## @param workers.embedding_worker.strategy.type Deployment strategy type
  ## @param workers.embedding_worker.scaling Horizontal Pod Autoscaler configuration
  ## @param workers.embedding_worker.env Environment variables for the embedding worker
  ## @param workers.embedding_worker.envFrom Environment variables from ConfigMap/Secret
  ## @skip workers.embedding_worker.resources.requests.cpu
  ## @skip workers.embedding_worker.resources.requests.memory
  ## @param workers.embedding_worker.serviceAccount.create Specifies whether a service account should be created
  ## @param workers.embedding_worker.serviceAccount.name The name of the service account to use
  ## @param workers.embedding_worker.serviceAccount.labels Additional labels for the service account
  ## @param workers.embedding_worker.serviceAccount.annotations Additional annotations for the service account
  ## @param workers.embedding_worker.enableHealthProbes Enable health probes for the embedding worker
  ## @param workers.embedding_worker.nodeSelector Node selector for embedding worker pods
  ## @param workers.embedding_worker.tolerations Tolerations for embedding worker pods
  ## @param workers.embedding_worker.affinity Affinity for embedding worker pods
  embedding_worker:
    enabled: true
    replicas: 1
    annotations: {}
    strategy:
      type: RollingUpdate
    scaling: {}
    # minReplicas: 1
    # maxReplicas: 5
    # metrics:
    #   cpu: 70
    #   memory: 70
    env: {}
    envFrom: []
    resources:
      requests:
        cpu: "4"
        memory: 8Gi
        #limits:
        #  cpu: "8"
        #  memory: 15Gi
    serviceAccount:
      create: false
      name: ""
      labels: {}
      annotations: {}
    enableHealthProbes: true
    nodeSelector: {}
    tolerations: []
    affinity: {}
  ## @param workers.summary_generation_worker.enabled Enable summary generation worker deployment
  ## @param workers.summary_generation_worker.replicas Number of summary generation worker replicas to deploy
  ## @param workers.summary_generation_worker.annotations Annotations for the summary generation worker deployment
  ## @param workers.summary_generation_worker.strategy.type Deployment strategy type
  ## @param workers.summary_generation_worker.scaling Horizontal Pod Autoscaler configuration
  ## @param workers.summary_generation_worker.env Environment variables for the summary generation worker
  ## @param workers.summary_generation_worker.envFrom Environment variables from ConfigMap/Secret
  ## @skip workers.summary_generation_worker.resources.requests.cpu
  ## @skip workers.summary_generation_worker.resources.requests.memory
  ## @param workers.summary_generation_worker.serviceAccount.create Specifies whether a service account should be created
  ## @param workers.summary_generation_worker.serviceAccount.name The name of the service account to use
  ## @param workers.summary_generation_worker.serviceAccount.labels Additional labels for the service account
  ## @param workers.summary_generation_worker.serviceAccount.annotations Additional annotations for the service account
  ## @param workers.summary_generation_worker.enableHealthProbes Enable health probes for the summary generation worker
  ## @param workers.summary_generation_worker.nodeSelector Node selector for summary generation worker pods
  ## @param workers.summary_generation_worker.tolerations Tolerations for summary generation worker pods
  ## @param workers.summary_generation_worker.affinity Affinity for summary generation worker pods
  summary_generation_worker:
    enabled: true
    replicas: 1
    annotations: {}
    strategy:
      type: RollingUpdate
    scaling: {}
    # minReplicas: 1
    # maxReplicas: 5
    # metrics:
    #   cpu: 70
    #   memory: 70
    env: {}
    envFrom: []
    resources:
      requests:
        cpu: "4"
        memory: 8Gi
        #limits:
        #  cpu: "8"
        #  memory: 15Gi
    serviceAccount:
      create: false
      name: ""
      labels: {}
      annotations: {}
    enableHealthProbes: true
    nodeSelector: {}
    tolerations: []
    affinity: {}
  ## @param workers.file_status_worker.enabled Enable summary generation worker deployment
  ## @param workers.file_status_worker.replicas Number of summary generation worker replicas to deploy
  ## @param workers.file_status_worker.annotations Annotations for the summary generation worker deployment
  ## @param workers.file_status_worker.strategy.type Deployment strategy type
  ## @param workers.file_status_worker.scaling Horizontal Pod Autoscaler configuration
  ## @param workers.file_status_worker.env Environment variables for the summary generation worker
  ## @param workers.file_status_worker.envFrom Environment variables from ConfigMap/Secret
  ## @skip workers.file_status_worker.resources.requests.cpu
  ## @skip workers.file_status_worker.resources.requests.memory
  ## @param workers.file_status_worker.serviceAccount.create Specifies whether a service account should be created
  ## @param workers.file_status_worker.serviceAccount.name The name of the service account to use
  ## @param workers.file_status_worker.serviceAccount.labels Additional labels for the service account
  ## @param workers.file_status_worker.serviceAccount.annotations Additional annotations for the service account
  ## @param workers.file_status_worker.enableHealthProbes Enable health probes for the summary generation worker
  ## @param workers.file_status_worker.nodeSelector Node selector for summary generation worker pods
  ## @param workers.file_status_worker.tolerations Tolerations for summary generation worker pods
  ## @param workers.file_status_worker.affinity Affinity for summary generation worker pods
  file_status_worker:
    enabled: true
    replicas: 1
    annotations: {}
    strategy:
      type: RollingUpdate
    scaling: {}
    # minReplicas: 1
    # maxReplicas: 5
    # metrics:
    #   cpu: 70
    #   memory: 70
    env: {}
    envFrom: []
    resources:
      requests:
        cpu: "1500m"
        memory: 2Gi
        #limits:
        #  cpu: "8"
        #  memory: 15Gi
    serviceAccount:
      create: false
      name: ""
      labels: {}
      annotations: {}
    enableHealthProbes: true
    nodeSelector: {}
    tolerations: []
    affinity: {}
  ## @param workers.skywell_worker.enabled Enable summary generation worker deployment
  ## @param workers.skywell_worker.replicas Number of summary generation worker replicas to deploy
  ## @param workers.skywell_worker.annotations Annotations for the summary generation worker deployment
  ## @param workers.skywell_worker.strategy.type Deployment strategy type
  ## @param workers.skywell_worker.scaling Horizontal Pod Autoscaler configuration
  ## @param workers.skywell_worker.env Environment variables for the summary generation worker
  ## @param workers.skywell_worker.envFrom Environment variables from ConfigMap/Secret
  ## @skip workers.skywell_worker.resources.requests.cpu
  ## @skip workers.skywell_worker.resources.requests.memory
  ## @param workers.skywell_worker.serviceAccount.create Specifies whether a service account should be created
  ## @param workers.skywell_worker.serviceAccount.name The name of the service account to use
  ## @param workers.skywell_worker.serviceAccount.labels Additional labels for the service account
  ## @param workers.skywell_worker.serviceAccount.annotations Additional annotations for the service account
  ## @param workers.skywell_worker.enableHealthProbes Enable health probes for the summary generation worker
  ## @param workers.skywell_worker.nodeSelector Node selector for summary generation worker pods
  ## @param workers.skywell_worker.tolerations Tolerations for summary generation worker pods
  ## @param workers.skywell_worker.affinity Affinity for summary generation worker pods
  skywell_worker:
    enabled: true
    replicas: 1
    annotations: {}
    strategy:
      type: RollingUpdate
    scaling: {}
    # minReplicas: 1
    # maxReplicas: 5
    # metrics:
    #   cpu: 70
    #   memory: 70
    env: {}
    envFrom: []
    resources:
      requests:
        cpu: "100m"
        memory: 2Gi
        #limits:
        #  cpu: "8"
        #  memory: 15Gi
    serviceAccount:
      create: false
      name: ""
      labels: {}
      annotations: {}
    enableHealthProbes: true
    nodeSelector: {}
    tolerations: []
    affinity: {}
## @section Models parameters
models:
  ## @param models[0].name Model name for the visual model
  ## @param models[0].load_models List of models to load
  ## @param models[0].replicas Number of visual model replicas to deploy
  ## @param models[0].port Port for the visual model service
  ## @param models[0].strategy.type Deployment strategy type
  ## @param models[0].scaling Horizontal Pod Autoscaler configuration
  ## @param models[0].image.repository Docker image repository for visual model
  ## @param models[0].service.type Kubernetes service type
  ## @param models[0].service.port Service port
  ## @param models[0].ingress.enabled Enable ingress record generation for visual model
  ## @param models[0].ingress.className IngressClass that will be used to implement the Ingress
  ## @param models[0].ingress.annotations Additional custom annotations for the ingress record
  ## @param models[0].ingress.host Default host for the ingress record
  ## @skip models[0].ingress.paths
  ## @param models[0].ingress.tls Enable TLS configuration for the hostname
  ## @param models[0].env Environment variables for the visual model
  ## @param models[0].envFrom Environment variables from ConfigMap/Secret
  ## @skip models[0].resources.requests.cpu
  ## @skip models[0].resources.requests.memory
  ## @skip models[0].resources.requests.nvidia.com/gpu
  ## @param models[0].serviceAccount.create Specifies whether a service account should be created
  ## @param models[0].serviceAccount.name The name of the service account to use
  ## @param models[0].serviceAccount.labels Additional labels for the service account
  ## @param models[0].serviceAccount.annotations Additional annotations for the service account
  ## @param models[0].enableHealthProbes Enable health probes for the visual model
  ## @param models[0].nodeSelector Node selector for visual model pods
  ## @param models[0].tolerations Tolerations for visual model pods
  ## @param models[0].affinity Affinity for visual model pods
  - name: models
    replicas: 1
    load_models:
      - colnomic # can be used for ingest/retrieval of images,  suggested to provide at least 16GiB RAM and A100 GPU
      # - inf-retriever-v1  # can be used for ingest/retrieval of texts,   suggested to provide at least 4GiB  RAM and A10/L4 GPU
      # - doclayout_parser  # can be used to generate images for answers,  suggested to provide at least 4GiB  RAM and A10/L4 GPU
    port: 8000
    strategy:
      type: RollingUpdate
    scaling: {}
    # minReplicas: 1
    # maxReplicas: 5
    # metrics:
    #   cpu: 70
    #   memory: 70
    image:
      repository: activeloopai/neohorizon-models-triton
    service:
      type: ClusterIP
      port: 80
    ingress:
      enabled: false
      className: ""
      annotations: {}
      host: "api.example.com"
      paths:
        - path: /
          pathType: Prefix
      tls: []
    env: {}
    envFrom: []
    resources:
      requests:
        cpu: "1"
        memory: 4Gi
        nvidia.com/gpu: "1"
        #limits:
        #  cpu: "2"
        #  memory: 16Gi
        #  nvidia.com/gpu: "1"
    serviceAccount:
      create: false
      name: ""
      labels: {}
      annotations: {}
    enableHealthProbes: false
    nodeSelector: {}
    tolerations: []
    affinity: {}
## @section Gotenberg parameters -- [source](https://github.com/MaikuMori/helm-charts/tree/master/charts/gotenberg)
## @param gotenberg.enabled Enable Gotenberg
## @param gotenberg.replicaCount Number of Gotenberg replicas
## @param gotenberg.api.timeout Gotenberg API timeout
## @skip gotenberg.resources.requests.cpu
## @skip gotenberg.resources.requests.memory
## @skip gotenberg.resources.limits.cpu
## @skip gotenberg.resources.limits.memory
gotenberg:
  enabled: true
  replicaCount: 1
  api:
    timeout: 120s
  resources:
    requests:
      cpu: 100m
      memory: 128Mi
    # limits:
    #   cpu: '2'
    #   memory: 4Gi