Jump to

• Request
• Response

(Preview) Update a model deployment

post

/20260430/aiDataPlatforms/{aiDataPlatformId}/mlops/api/2.0/model-deployments/update

(Preview) Updates a model deployment.

Request

Path Parameters

• aiDataPlatformId(required): string
  Minimum Length: 1

  Maximum Length: 500

  The [OCID](/iaas/Content/General/Concepts/identifiers.htm) of the AI Data Platform (Data Lake) instance.

Header Parameters

• opc-request-id: string
  Maximum Length: 255

  Unique Oracle-assigned identifier for the request. If you need to contact Oracle about a particular request, please provide the request ID. The only valid characters for request IDs are letters, numbers, underscore, and dash.

Body ()

Details for the model deployment update.

Root Schema : UpdateModelDeploymentDetails

Type: object

The data to update a model deployment. The registered model and model version cannot be changed here; use the roll-forward and roll-back actions instead.

Show Source

• authDetails: object DeploymentOAuthDetails
  OAuth trust configuration for a model deployment's query endpoint. Required when authType is OAUTH and must be omitted for AIDP. A closed schema: the gateway turns these values into the token policy it enforces, so a member outside this set is not accepted -- it is discarded rather than stored or forwarded, and a misspelled member therefore leaves the one it was meant to be unset, which fails validation. The serialized object must not exceed 16000 characters. Cannot be changed while the deployment is ACTIVE.
• authType: string
  Allowed Values: [ "AIDP", "OAUTH" ]

  Authentication mechanism for the deployment's query endpoint. Required when authDetails is provided. Cannot be changed while the deployment is ACTIVE.
• computeKey(required): string
  Minimum Length: 1

  Maximum Length: 255

  Compute key of the deployment. At most 255 characters and 255 UTF-8 bytes.
• deployment_id(required): string
  Minimum Length: 32

  Maximum Length: 32

  Pattern: ^[0-9a-f]{32}$

  ID of the deployment.
• deployment_targets: array deployment_targets
  Maximum Number of Items: 100

  Model versions served by the deployment. Immutable through update (use roll-forward/roll-back); accepted here only when unchanged.
• description: string
  Maximum Length: 2000

  Description of the deployment. At most 2000 characters and 2000 UTF-8 bytes.
• name(required): string
  Minimum Length: 1

  Maximum Length: 255

  Name of the deployment. At most 255 characters and 255 UTF-8 bytes.
• workspaceKey(required): string
  Minimum Length: 1

  Maximum Length: 255

  Workspace key of the deployment. At most 255 characters and 255 UTF-8 bytes.

{
    "description":"The data to update a model deployment. The registered model and model version cannot be changed here; use the roll-forward and roll-back actions instead.",
    "type":"object",
    "required":[
        "deployment_id",
        "name",
        "workspaceKey",
        "computeKey"
    ],
    "properties":{
        "deployment_id":{
            "description":"ID of the deployment.",
            "type":"string",
            "minLength":32,
            "maxLength":32,
            "pattern":"^[0-9a-f]{32}$"
        },
        "name":{
            "description":"Name of the deployment. At most 255 characters and 255 UTF-8 bytes.",
            "type":"string",
            "minLength":1,
            "maxLength":255
        },
        "description":{
            "description":"Description of the deployment. At most 2000 characters and 2000 UTF-8 bytes.",
            "type":"string",
            "maxLength":2000
        },
        "workspaceKey":{
            "description":"Workspace key of the deployment. At most 255 characters and 255 UTF-8 bytes.",
            "type":"string",
            "minLength":1,
            "maxLength":255
        },
        "computeKey":{
            "description":"Compute key of the deployment. At most 255 characters and 255 UTF-8 bytes.",
            "type":"string",
            "minLength":1,
            "maxLength":255
        },
        "deployment_targets":{
            "description":"Model versions served by the deployment. Immutable through update (use roll-forward/roll-back); accepted here only when unchanged.",
            "type":"array",
            "maxItems":100,
            "items":{
                "$ref":"#/definitions/DeploymentTarget"
            }
        },
        "authType":{
            "description":"Authentication mechanism for the deployment's query endpoint. Required when authDetails is provided. Cannot be changed while the deployment is ACTIVE.",
            "type":"string",
            "x-obmcs-top-level-enum":"#/definitions/DeploymentAuthType",
            "enum":[
                "AIDP",
                "OAUTH"
            ]
        },
        "authDetails":{
            "$ref":"#/definitions/DeploymentOAuthDetails"
        }
    }
}

Nested Schema : DeploymentOAuthDetails

Type: object

OAuth trust configuration for a model deployment's query endpoint. Required when authType is OAUTH and must be omitted for AIDP. A closed schema: the gateway turns these values into the token policy it enforces, so a member outside this set is not accepted -- it is discarded rather than stored or forwarded, and a misspelled member therefore leaves the one it was meant to be unset, which fails validation. The serialized object must not exceed 16000 characters. Cannot be changed while the deployment is ACTIVE.

Show Source

• audienceClaim(required): array audienceClaim
  Minimum Number of Items: 1

  Maximum Number of Items: 20

  OAuth audience claims (aud) the query endpoint accepts. Entries must be unique.
• issuerClaim(required): string
  Minimum Length: 1

  Maximum Length: 1024

  OAuth issuer claim (iss) the query endpoint requires. An https URI whose host is a fully qualified domain name, not an IP literal in any notation, and which carries no userinfo and no fragment.
• jwksUri(required): string
  Minimum Length: 1

  Maximum Length: 2048

  URI the gateway retrieves the signing keys (JWKS) from. An https URI whose host is a fully qualified domain name, not an IP literal in any notation, and which carries no userinfo and no fragment. A query string and a non-default port are permitted: published provider JWKS endpoints use both.

{
    "description":"OAuth trust configuration for a model deployment's query endpoint. Required when authType is OAUTH and must be omitted for AIDP. A closed schema: the gateway turns these values into the token policy it enforces, so a member outside this set is not accepted -- it is discarded rather than stored or forwarded, and a misspelled member therefore leaves the one it was meant to be unset, which fails validation. The serialized object must not exceed 16000 characters. Cannot be changed while the deployment is ACTIVE.",
    "type":"object",
    "additionalProperties":false,
    "required":[
        "issuerClaim",
        "audienceClaim",
        "jwksUri"
    ],
    "properties":{
        "issuerClaim":{
            "description":"OAuth issuer claim (iss) the query endpoint requires. An https URI whose host is a fully qualified domain name, not an IP literal in any notation, and which carries no userinfo and no fragment.",
            "type":"string",
            "minLength":1,
            "maxLength":1024
        },
        "audienceClaim":{
            "description":"OAuth audience claims (aud) the query endpoint accepts. Entries must be unique.",
            "type":"array",
            "minItems":1,
            "maxItems":20,
            "items":{
                "type":"string",
                "minLength":1,
                "maxLength":1024
            }
        },
        "jwksUri":{
            "description":"URI the gateway retrieves the signing keys (JWKS) from. An https URI whose host is a fully qualified domain name, not an IP literal in any notation, and which carries no userinfo and no fragment. A query string and a non-default port are permitted: published provider JWKS endpoints use both.",
            "type":"string",
            "minLength":1,
            "maxLength":2048
        }
    }
}

Nested Schema : deployment_targets

Type: array

Maximum Number of Items: 100

Model versions served by the deployment. Immutable through update (use roll-forward/roll-back); accepted here only when unchanged.

Show Source

• Array of: object DeploymentTarget
  A model version targeted by a deployment and its traffic share.

{
    "description":"Model versions served by the deployment. Immutable through update (use roll-forward/roll-back); accepted here only when unchanged.",
    "type":"array",
    "maxItems":100,
    "items":{
        "$ref":"#/definitions/DeploymentTarget"
    }
}

Nested Schema : audienceClaim

Type: array

Minimum Number of Items: 1

Maximum Number of Items: 20

OAuth audience claims (aud) the query endpoint accepts. Entries must be unique.

Show Source

• Array of: string
  Minimum Length: 1

  Maximum Length: 1024

{
    "description":"OAuth audience claims (aud) the query endpoint accepts. Entries must be unique.",
    "type":"array",
    "minItems":1,
    "maxItems":20,
    "items":{
        "type":"string",
        "minLength":1,
        "maxLength":1024
    }
}

Nested Schema : DeploymentTarget

Type: object

A model version targeted by a deployment and its traffic share.

Show Source

• model_version(required): string
  Minimum Length: 1

  Maximum Length: 30

  Version number of the model version.
• traffic_percentage(required): integer
  Minimum Value: 1

  Maximum Value: 100

  Percentage of deployment traffic routed to this model version.

{
    "description":"A model version targeted by a deployment and its traffic share.",
    "type":"object",
    "required":[
        "model_version",
        "traffic_percentage"
    ],
    "properties":{
        "model_version":{
            "description":"Version number of the model version.",
            "type":"string",
            "minLength":1,
            "maxLength":30
        },
        "traffic_percentage":{
            "description":"Percentage of deployment traffic routed to this model version.",
            "type":"integer",
            "minimum":1,
            "maximum":100
        }
    }
}

Back to Top

Response

Supported Media Types

• application/json

200 Response

Successful operation. Model deployment is updated.

Headers

• etag: string
  For optimistic concurrency control. See `if-match`.
• opc-request-id: string
  Unique Oracle-assigned ID for the request. If you need to contact Oracle about a particular request, please provide the request ID.

Body ()

Root Schema : ModelDeployment

Type: object

A model deployment.

Show Source

• activated_by: string
  User that activated the model deployment.
• activated_time: integer(int64)
  Unix timestamp in milliseconds of when the deployment was activated.
• authDetails: object DeploymentOAuthDetails
  OAuth trust configuration for a model deployment's query endpoint. Required when authType is OAUTH and must be omitted for AIDP. A closed schema: the gateway turns these values into the token policy it enforces, so a member outside this set is not accepted -- it is discarded rather than stored or forwarded, and a misspelled member therefore leaves the one it was meant to be unset, which fails validation. The serialized object must not exceed 16000 characters. Cannot be changed while the deployment is ACTIVE.
• authType: string
  Allowed Values: [ "AIDP", "OAUTH" ]

  Authentication mechanism selected for the deployment's query endpoint.
• computeKey(required): string
  Compute key of the deployment.
• created_by(required): string
  User that created the model deployment.
• created_time(required): string
  Unix timestamp in milliseconds of when the deployment was created.
• deployment_id(required): string
  ID of the deployment.
• deployment_targets(required): array deployment_targets
  Deployment targets of the deployment.
• description: string
  Description of the deployment.
• model_name(required): string
  Name of the registered model.
• name(required): string
  Name of the deployment.
• serving_uri: string
  Maximum Length: 1000

  Serving URI of the deployment.
• status(required): string
  Allowed Values: [ "INACTIVE", "ACTIVATING", "ACTIVE", "DEACTIVATING", "UPDATING" ]

  Status of the deployment.
• tags: array tags
  List of tags set on the model deployment.
• updated_by(required): string
  User that last updated the model deployment.
• updated_time(required): string
  Unix timestamp in milliseconds of when the deployment was updated.
• workspaceKey(required): string
  Workspace key of the deployment.

{
    "description":"A model deployment.",
    "type":"object",
    "required":[
        "deployment_id",
        "name",
        "model_name",
        "deployment_targets",
        "workspaceKey",
        "computeKey",
        "status",
        "created_time",
        "updated_time",
        "created_by",
        "updated_by"
    ],
    "properties":{
        "deployment_id":{
            "description":"ID of the deployment.",
            "type":"string"
        },
        "name":{
            "description":"Name of the deployment.",
            "type":"string"
        },
        "description":{
            "description":"Description of the deployment.",
            "type":"string"
        },
        "model_name":{
            "description":"Name of the registered model.",
            "type":"string"
        },
        "deployment_targets":{
            "description":"Deployment targets of the deployment.",
            "type":"array",
            "items":{
                "$ref":"#/definitions/DeploymentTarget"
            }
        },
        "workspaceKey":{
            "description":"Workspace key of the deployment.",
            "type":"string"
        },
        "computeKey":{
            "description":"Compute key of the deployment.",
            "type":"string"
        },
        "status":{
            "description":"Status of the deployment.",
            "type":"string",
            "x-obmcs-top-level-enum":"#/definitions/DeploymentStatus",
            "enum":[
                "INACTIVE",
                "ACTIVATING",
                "ACTIVE",
                "DEACTIVATING",
                "UPDATING"
            ]
        },
        "activated_time":{
            "description":"Unix timestamp in milliseconds of when the deployment was activated.",
            "type":"integer",
            "format":"int64"
        },
        "activated_by":{
            "description":"User that activated the model deployment.",
            "type":"string"
        },
        "serving_uri":{
            "description":"Serving URI of the deployment.",
            "type":"string",
            "maxLength":1000
        },
        "created_time":{
            "description":"Unix timestamp in milliseconds of when the deployment was created.",
            "type":"string"
        },
        "updated_time":{
            "description":"Unix timestamp in milliseconds of when the deployment was updated.",
            "type":"string"
        },
        "created_by":{
            "description":"User that created the model deployment.",
            "type":"string"
        },
        "updated_by":{
            "description":"User that last updated the model deployment.",
            "type":"string"
        },
        "tags":{
            "description":"List of tags set on the model deployment.",
            "type":"array",
            "items":{
                "$ref":"#/definitions/ModelDeploymentTag"
            }
        },
        "authType":{
            "description":"Authentication mechanism selected for the deployment's query endpoint.",
            "type":"string",
            "x-obmcs-top-level-enum":"#/definitions/DeploymentAuthType",
            "enum":[
                "AIDP",
                "OAUTH"
            ]
        },
        "authDetails":{
            "$ref":"#/definitions/DeploymentOAuthDetails"
        }
    }
}

Nested Schema : DeploymentOAuthDetails

Type: object

OAuth trust configuration for a model deployment's query endpoint. Required when authType is OAUTH and must be omitted for AIDP. A closed schema: the gateway turns these values into the token policy it enforces, so a member outside this set is not accepted -- it is discarded rather than stored or forwarded, and a misspelled member therefore leaves the one it was meant to be unset, which fails validation. The serialized object must not exceed 16000 characters. Cannot be changed while the deployment is ACTIVE.

Show Source

• audienceClaim(required): array audienceClaim
  Minimum Number of Items: 1

  Maximum Number of Items: 20

  OAuth audience claims (aud) the query endpoint accepts. Entries must be unique.
• issuerClaim(required): string
  Minimum Length: 1

  Maximum Length: 1024

  OAuth issuer claim (iss) the query endpoint requires. An https URI whose host is a fully qualified domain name, not an IP literal in any notation, and which carries no userinfo and no fragment.
• jwksUri(required): string
  Minimum Length: 1

  Maximum Length: 2048

  URI the gateway retrieves the signing keys (JWKS) from. An https URI whose host is a fully qualified domain name, not an IP literal in any notation, and which carries no userinfo and no fragment. A query string and a non-default port are permitted: published provider JWKS endpoints use both.

{
    "description":"OAuth trust configuration for a model deployment's query endpoint. Required when authType is OAUTH and must be omitted for AIDP. A closed schema: the gateway turns these values into the token policy it enforces, so a member outside this set is not accepted -- it is discarded rather than stored or forwarded, and a misspelled member therefore leaves the one it was meant to be unset, which fails validation. The serialized object must not exceed 16000 characters. Cannot be changed while the deployment is ACTIVE.",
    "type":"object",
    "additionalProperties":false,
    "required":[
        "issuerClaim",
        "audienceClaim",
        "jwksUri"
    ],
    "properties":{
        "issuerClaim":{
            "description":"OAuth issuer claim (iss) the query endpoint requires. An https URI whose host is a fully qualified domain name, not an IP literal in any notation, and which carries no userinfo and no fragment.",
            "type":"string",
            "minLength":1,
            "maxLength":1024
        },
        "audienceClaim":{
            "description":"OAuth audience claims (aud) the query endpoint accepts. Entries must be unique.",
            "type":"array",
            "minItems":1,
            "maxItems":20,
            "items":{
                "type":"string",
                "minLength":1,
                "maxLength":1024
            }
        },
        "jwksUri":{
            "description":"URI the gateway retrieves the signing keys (JWKS) from. An https URI whose host is a fully qualified domain name, not an IP literal in any notation, and which carries no userinfo and no fragment. A query string and a non-default port are permitted: published provider JWKS endpoints use both.",
            "type":"string",
            "minLength":1,
            "maxLength":2048
        }
    }
}

Nested Schema : deployment_targets

Type: array

Deployment targets of the deployment.

Show Source

• Array of: object DeploymentTarget
  A model version targeted by a deployment and its traffic share.

{
    "description":"Deployment targets of the deployment.",
    "type":"array",
    "items":{
        "$ref":"#/definitions/DeploymentTarget"
    }
}

Nested Schema : tags

Type: array

List of tags set on the model deployment.

Show Source

• Array of: object ModelDeploymentTag
  A tag associated with a model deployment.

{
    "description":"List of tags set on the model deployment.",
    "type":"array",
    "items":{
        "$ref":"#/definitions/ModelDeploymentTag"
    }
}

Nested Schema : audienceClaim

Type: array

Minimum Number of Items: 1

Maximum Number of Items: 20

OAuth audience claims (aud) the query endpoint accepts. Entries must be unique.

Show Source

• Array of: string
  Minimum Length: 1

  Maximum Length: 1024

{
    "description":"OAuth audience claims (aud) the query endpoint accepts. Entries must be unique.",
    "type":"array",
    "minItems":1,
    "maxItems":20,
    "items":{
        "type":"string",
        "minLength":1,
        "maxLength":1024
    }
}

Nested Schema : DeploymentTarget

Type: object

A model version targeted by a deployment and its traffic share.

Show Source

• model_version(required): string
  Minimum Length: 1

  Maximum Length: 30

  Version number of the model version.
• traffic_percentage(required): integer
  Minimum Value: 1

  Maximum Value: 100

  Percentage of deployment traffic routed to this model version.

{
    "description":"A model version targeted by a deployment and its traffic share.",
    "type":"object",
    "required":[
        "model_version",
        "traffic_percentage"
    ],
    "properties":{
        "model_version":{
            "description":"Version number of the model version.",
            "type":"string",
            "minLength":1,
            "maxLength":30
        },
        "traffic_percentage":{
            "description":"Percentage of deployment traffic routed to this model version.",
            "type":"integer",
            "minimum":1,
            "maximum":100
        }
    }
}

Nested Schema : ModelDeploymentTag

Type: object

A tag associated with a model deployment.

Show Source

• key(required): string
  Minimum Length: 1

  Maximum Length: 250

  Key of the model deployment tag. At most 250 characters and 250 UTF-8 bytes.
• value: string
  Maximum Length: 5000

  Value of the model deployment tag. At most 5000 characters and 5000 UTF-8 bytes.

{
    "description":"A tag associated with a model deployment.",
    "type":"object",
    "required":[
        "key"
    ],
    "properties":{
        "key":{
            "type":"string",
            "minLength":1,
            "maxLength":250,
            "description":"Key of the model deployment tag. At most 250 characters and 250 UTF-8 bytes."
        },
        "value":{
            "type":"string",
            "maxLength":5000,
            "description":"Value of the model deployment tag. At most 5000 characters and 5000 UTF-8 bytes."
        }
    }
}

400 Response

Bad Request (invalid query parameters, malformed headers, and so on).

Headers

• opc-request-id: string
  Unique Oracle-assigned ID for the request. If you need to contact Oracle about a particular request, please provide the request ID.

Body ()

Root Schema : Error

Type: object

Error information.

Show Source

• code(required): string
  A short error code that defines the error, meant for programmatic parsing.
• message(required): string
  A human-readable error message.

{
    "description":"Error information.",
    "required":[
        "code",
        "message"
    ],
    "properties":{
        "code":{
            "description":"A short error code that defines the error, meant for programmatic parsing.",
            "type":"string"
        },
        "message":{
            "description":"A human-readable error message.",
            "type":"string"
        }
    }
}

401 Response

Unauthorized (missing or expired credentials, and so on).

Headers

• opc-request-id: string
  Unique Oracle-assigned ID for the request. If you need to contact Oracle about a particular request, please provide the request ID.

Body ()

Root Schema : Error

Type: object

Error information.

Show Source

• code(required): string
  A short error code that defines the error, meant for programmatic parsing.
• message(required): string
  A human-readable error message.

{
    "description":"Error information.",
    "required":[
        "code",
        "message"
    ],
    "properties":{
        "code":{
            "description":"A short error code that defines the error, meant for programmatic parsing.",
            "type":"string"
        },
        "message":{
            "description":"A human-readable error message.",
            "type":"string"
        }
    }
}

404 Response

Not Found. The requested resource was not found.

Headers

• opc-request-id: string
  Unique Oracle-assigned ID for the request. If you need to contact Oracle about a particular request, please provide the request ID.

Body ()

Root Schema : Error

Type: object

Error information.

Show Source

• code(required): string
  A short error code that defines the error, meant for programmatic parsing.
• message(required): string
  A human-readable error message.

{
    "description":"Error information.",
    "required":[
        "code",
        "message"
    ],
    "properties":{
        "code":{
            "description":"A short error code that defines the error, meant for programmatic parsing.",
            "type":"string"
        },
        "message":{
            "description":"A human-readable error message.",
            "type":"string"
        }
    }
}

409 Response

Conflict. Request conflicts with the current state of the resource.

Headers

• opc-request-id: string
  Unique Oracle-assigned ID for the request. If you need to contact Oracle about a particular request, please provide the request ID.

Body ()

Root Schema : Error

Type: object

Error information.

Show Source

• code(required): string
  A short error code that defines the error, meant for programmatic parsing.
• message(required): string
  A human-readable error message.

{
    "description":"Error information.",
    "required":[
        "code",
        "message"
    ],
    "properties":{
        "code":{
            "description":"A short error code that defines the error, meant for programmatic parsing.",
            "type":"string"
        },
        "message":{
            "description":"A human-readable error message.",
            "type":"string"
        }
    }
}

429 Response

Too Many Requests. Too many requests sent to the server in a short period.

Headers

• opc-request-id: string
  Unique Oracle-assigned ID for the request. If you need to contact Oracle about a particular request, please provide the request ID.

Body ()

Root Schema : Error

Type: object

Error information.

Show Source

• code(required): string
  A short error code that defines the error, meant for programmatic parsing.
• message(required): string
  A human-readable error message.

{
    "description":"Error information.",
    "required":[
        "code",
        "message"
    ],
    "properties":{
        "code":{
            "description":"A short error code that defines the error, meant for programmatic parsing.",
            "type":"string"
        },
        "message":{
            "description":"A human-readable error message.",
            "type":"string"
        }
    }
}

500 Response

Internal Server Error. The server encountered an unexpected condition preventing fulfilment of the request.

Headers

• opc-request-id: string
  Unique Oracle-assigned ID for the request. If you need to contact Oracle about a particular request, please provide the request ID.

Body ()

Root Schema : Error

Type: object

Error information.

Show Source

• code(required): string
  A short error code that defines the error, meant for programmatic parsing.
• message(required): string
  A human-readable error message.

{
    "description":"Error information.",
    "required":[
        "code",
        "message"
    ],
    "properties":{
        "code":{
            "description":"A short error code that defines the error, meant for programmatic parsing.",
            "type":"string"
        },
        "message":{
            "description":"A human-readable error message.",
            "type":"string"
        }
    }
}

Default Response

Unknown Error. Error is not recognized by the system.

Headers

• opc-request-id: string
  Unique Oracle-assigned ID for the request. If you need to contact Oracle about a particular request, please provide the request ID.

Body ()

Root Schema : Error

Type: object

Error information.

Show Source

• code(required): string
  A short error code that defines the error, meant for programmatic parsing.
• message(required): string
  A human-readable error message.

{
    "description":"Error information.",
    "required":[
        "code",
        "message"
    ],
    "properties":{
        "code":{
            "description":"A short error code that defines the error, meant for programmatic parsing.",
            "type":"string"
        },
        "message":{
            "description":"A human-readable error message.",
            "type":"string"
        }
    }
}

Back to Top