UnifiedWriteInferenceV3DeploymentRequest.md

UnifiedWriteInferenceV3DeploymentRequest

Properties

Name	Type	Description	Notes
max_surge	int		[optional]
max_unavailable	int		[optional]
name	str
cluster_id	int
hardware_instance_id	int
user_annotations	Dict[str, str]		[optional]
chart_revision	str		[optional]
volume_mounts	List[VolumeMount]		[optional]
priority	str		[optional]
hf_token_ref	int		[optional]
image_url	str
image_pull_secret_credentials	ImagePullSecretCredentials		[optional]
image_pull_ref	int		[optional]
image_pull_secret_name	str	Name of a pre-existing image pull Secret in the deployment namespace. Created, refreshed and deleted externally. Mutually exclusive with inline registry credentials and image_pull_ref. Include on updates to retain it; null clears it.	[optional]
port	int
min_replicas	int
max_replicas	int
concurrency	int		[optional]
cooldown_period	int		[optional]
healthcheck	str		[optional]
env_vars	Dict[str, str]		[optional]
command	str		[optional]
endpoint_bearer_token	str		[optional]
bearer_token_ref	int		[optional]
endpoint_certificate_authority	str		[optional]
hf_token	str		[optional]
env_vars_refs	object		[optional]
backend_protocol	BackendProtocol		[optional]
enable_logging	bool		[optional] [default to False]
enable_node_model_cache	bool		[optional] [default to False]
session_affinity	bool	Enable best-effort sticky routing via the `X-Session-Id` request header. Requests carrying the same header value land on the same pod, improving KV cache reuse for agentic workloads. Requests without the header are routed at random. Affinity is NOT durable: scaling, rollouts, restarts, or readiness-probe transitions will remap sessions to different pods. Do not use for irreplaceable in-pod state.	[optional] [default to False]
config_file	ConfigFileMount		[optional]
metrics	MetricsConfig		[optional]
type	str

Example

from platform_api_python_client.models.unified_write_inference_v3_deployment_request import UnifiedWriteInferenceV3DeploymentRequest

# TODO update the JSON string below
json = "{}"
# create an instance of UnifiedWriteInferenceV3DeploymentRequest from a JSON string
unified_write_inference_v3_deployment_request_instance = UnifiedWriteInferenceV3DeploymentRequest.from_json(json)
# print the JSON string representation of the object
print(UnifiedWriteInferenceV3DeploymentRequest.to_json())

# convert the object into a dict
unified_write_inference_v3_deployment_request_dict = unified_write_inference_v3_deployment_request_instance.to_dict()
# create an instance of UnifiedWriteInferenceV3DeploymentRequest from a dict
unified_write_inference_v3_deployment_request_from_dict = UnifiedWriteInferenceV3DeploymentRequest.from_dict(unified_write_inference_v3_deployment_request_dict)

[Back to Model list] [Back to API list] [Back to README]