Skip to main content
POST
/
v4
/
serving
Python
import os
from scale_gp import SGPClient

client = SGPClient(
    api_key=os.environ.get("SGP_API_KEY"),  # This is the default and can be omitted
)
model_server_info = client.model_servers.create(
    model_deployment_id="model_deployment_id",
    name="name",
)
print(model_server_info.account_id)
{
  "name": "<string>",
  "model_server_id": "<string>",
  "account_id": "<string>",
  "model_deployment_id": "<string>",
  "alias": "<string>"
}

Authorizations

​
x-api-key
string
header
required

Headers

​
x-selected-account-id
string | null

Body

application/json

Create a new model server with an initial backend configuration.

Args: name (str): The name of the model server. Only used for display purposes. model_deployment_id (str): The ID of the model deployment to use as the initial backend. alias (str, optional): An alias for the model server. Defaults to None. If configured the model server can be accessed via the alias instead of the model server ID. i.e /models/server/alias/{alias}/execute

​
name
string
required
​
model_deployment_id
string
required
​
alias
string

Response

Successful Response

Model server information. name: The name of the model server. Only used for display purposes. model_server_id: The ID of the model server. model_deployment_id: The ID of the model deployment being used as the backend. account_id: The ID of the account that owns the model server. alias: An alias for the model server. If configured the model server can be accessed via the alias instead of the model server ID. i.e /models/server/alias/{alias}/execute

​
name
string
required
​
model_server_id
string
required
​
account_id
string
required
​
model_deployment_id
string
​
alias
string