I have an MLflow model deployed in SageMaker, but when I try to update it I get this error:

MlflowException: You are attempting to deploy an application with name: titanic-xgboost-endpoint in 'create' mode. However, an application with the same name already exists. If you want to update this application, deploy in 'add' or 'replace' mode.

However, get_deploy_client("sagemaker").create_deployment does not have a mode argument. Does anyone know what I am supposed to do here if I want to replace?

My deployment code is below:

client.create_deployment(
    name=ENDPOINT_NAME,
    model_uri=f"models:/{REGISTERED_MODEL_NAME}/latest",
    flavor="python_function",
    config={
        "region_name": AWS_REGION,
        "execution_role_arn": SAGEMAKER_ROLE_ARN,
        "instance_type": INSTANCE_TYPE,
        "instance_count": INSTANCE_COUNT,
        "synchronous": True,
        "timeout_seconds": 1800,
    }
)